Full text of "The Journal of Hellenic studies"

See other formats

[£!;

^ft.V;'"; : :

Hi'!

'•\'^ : y

3"iZtireA ^6*?*

THE JOURNAL

HELLENIC STUDIES

THE SOCIETY FOR THE PROMOTION OF HELLENIC STUDIES

THE JOURNAL

HELLENIC STUDIES

VOLUME XXVII. (1907

PUBLISHED BY THE COUNCIL, AND SOLD ON THEIR BEHALF

MACMILLAN AND CO., Limited, ST. MARTIN'S STREET

LONDON. W.C.

MDCCCCVII

The Rights of Translation and Reproduction are Reserved

Richard Clay and Sons, Limited

bread street hill, e.c. and

bungay, suffolk.

CONTENTS.

Rules of the Society

List of Officers and Members

Proceedings of the Society, 1906-1907

Financial Statement -

Additions to the Library

Accessions to the Catalogue of Slides

Notice to Contributors

Compton (W. C.) and H. Awdry Two Notes on Pylos and Sphacteria

xin
xix
... xliii
lx
... Ixv
... lxxii
. . . lxxvii
... 274

Dawkins (Pi. M.)
FORSTER (E. S.) ...
FOTHER INGHAM (J. K.)

Gardiner (E. N.)

Hasluck (F. W.
Hicks (E. L.) ..
Macdonald (G.)

Miller (W.) ..
Myres (J. L.) ..

Smith (C.)
Strzygowski (J.)
Tarn (W. W.) ..

Archaeology in Greece (1906-1907) ... 284
Terracottas from Boeotia and Crete ... 68
On the 'List of Thalassocracies ' in

Eusebius ...

Throwing the Diskos [Plates I. -III.] ...
Throwing the Javelin [Plates XVII.-

XX.]

Inscriptions from the Cyzicus District ...
Three Inscriptions from Asia Minor
Early Seleucid Portraits [Plates XIII.,

XIV.]

Monemvasia [Plates XV., XVI] ... 229, 300
The ' List of Thalassocracies * in Eusebius :

a Reply

A History of the Pelasgian Theory

The Central Groups of the Parthenon

Pediments

A Sarcophagus of the Sidamara type

[Plates V.-XIL]

The Fleet* of the First Punic War

75
1

249

226

145

123
170

212

99
48

CONTENTS.

TlLLYARD (H. J. W.) ...

Wells (J.)
Wroth (W. W.)

Notices of Books

Index of Subjects

Greek Index

List of Books Noticed

Instrumental Music in the Roman Age ... 160

The Persian Friends of Herodotus ... 37

Peparethus and its Coinage [Plate IV.] 90

131,302

315

321

322

CONTENTS.

LIST OF PLATES.

I. B.-F. Amphora in the British Museum (B 271).
II. Attic B.-F. Lekythos in the British Museum (B 576).

III. R.-F. Pelike in the British Museum (E 395).
B.-F. Fragment at Wurzburg.

IV. Coins of Peparethus, etc.

V. Figure B from the Cook Sarcophagus.

VI.

VII.

VIII.

IX.

)»

XI.

XII.

XIII, XIV. Early Seleucid Portraits.

XV. Monemvasia : Ilavayia MvpriSicoTicro-a — 'Ayi'a 2o<£ia.

XVI. ,, : Town Walls and Gate — Modern town at Base of

cliff.

XVII. Nolan Amphora in the British Museum (E 326).

XVIII. Panathenaic Amphora in the British Museum (B 131).

XIX. B.-F. Amphora in the British Museum (E 256).

XX. Panathenaic Vase in the British Museum.

CONTENTS.

LIST OF ILLUSTRATIONS IN THE TEXT.

Throwing 1 the Diskos.

1. R.-F. Kylix (Louvre G 96)

2. R.-F. Kylikes (after Jiithner)

3. R.-F. Kylix (B.M. E 6)

4. B.-F. Kelebe(B.M. B 361)

5. R.-F. Krater (after Hartwig)

6. Diskobolos on Bronze Lebes (B.M. 559)

7. Fifth Century Bronze (after Burl. Fine Arts Club Catal.)

8. From B.-F. Tripod (Berlin 1727)

9. 10. R.-F. Kylix (Naples : after Arch. Zeit.)

11. Bronze Diskobolos (B.M. 675)

12. Lekythos from Eretria ('E<£. 'Apx-)

13. The Standing Diskobolos (Vatican)

14. R.-F. Kylix (Munich 795)

15. R.-F. Kylix (from Gerhard)

16. R-F. Kylix (interior of Fig. 14)

17. R.-F. Kylix (after Hartwig)

18. Myron's Diskobolos (Lancelotti)

19. Coins of Cos in British Museum...

20. From Panathenaic Vase (Naples : after Bull. Nap.)

21. R.-F. Hydria (B.M. E 164 : after B.C.H.)

22. Athenian Lekythos (Boulogne : from Le Musee)

23. R.-F. Kylix (Boulogne : from Le Musee)

24. B.-F. Hydria (Vienna)

2
11
13
15

16
17
18
19
20
22
23
25
26
2G
27
27
28
30
32
32
33
34
34

Inscriptions from the Cyzieus District.

Inscription at Kermasti

Terracottas from Boeotia and Crete.

1. Primitive Standing Figure from Boeotia

2. Primitive Seated Figure ,. ,,

3. Primitive Standing Figure ..

4. Chariot Group from Crete

5. Horseman carrying Faggots

69
70
71
72
73

CONTENTS.

Peparethus and its Coinage.

A, B. Coins with winged figure and head of Herakles ...
C. Coin with seated Dionysos

9.
10.
11.
12.
13.
14.
15.
16.

A Sarcophagus of the Sidamara

Fragment A of the Cook Sarcophagus

The Sidamara Sarcophagus

Sarcophagus Fragments at Smyrna

Dioscurus from the Sidamara Sarcophagus
Sarcophagus from. Selefkeh at Constantinople ...

Development of the Sarcophagus-Capitals

Relief at Berlin

Sarcophagus Fragment in the British Museum

Muse from the Mantinean Basis

The 'Matron of Herculaneum ' (Dresden)

Niche in Cemetery of the Tulunids, near Cairo
From the Throne of St. Maximian, Bavenna . . .

Leaf of Ivory Diptych in the B.M

Portion of Pompeian Wall-Painting

Reconstruction of Pompeian Stage-Facade
Pompeian Wall-Painting

type.

Instrumental Music in the Roman Age.

1. Trigonum or Sambuca

2. Terracotta Figure at Susa (Woman with Zither)

3. Pandora

4. Terracotta Figure at Susa (Woman with Pandura and Wreath)

5. ,, ,, in Mus. Alaoui (Man playing on Pandura)

6. Sarcophagus in the Taormina Museum

7. Pan-pipes

Three Inscriptions from Asia Minor.

Inscription from Troy

Monemvasia.

1. Monemvasia from the Land

2. Entrance to Kastro

3. Kastro

4. Arms on Well- Head in the Castle

The Central Groups of the Pediments of the Parthenon.

1. Scheme of Restoration of Central Group of E. Pediment

2. ,, ,, „ ., W. Pediment

3. i:.-F. Vase with Birth of Athena

95
97

100
101
103
104
105
108
109
110
111
112
114
116
117
119
120
121

162
163
164
165
165
166
167

226

229
230
238
240

244
245
247

CONTENTS.

Throwing- the Javelin.

1. Various methods of attaching the amentum

2. Interior of B.-F. Kylix (B.M. 380)

3. From the Francois Vase (from Furtwangler)

4. U.se of the throwing-thong

5. E.-F. Psykter of Phintias (from Ant. Denkm.)

6. Panathenaic Amphora (Leyden, from Arch. Zeit.)

7. From R.-F. Stamnos (Mus. Greg.)

8. From B.-F. Vase from Acropolis, Athens

9. R.-F. Kylix (Munich: after Juthner)

10. R.-F. Kylix (Berlin: after Hartwig)

11. From R.-F. Kylix (Torlonia : after Juthner) .

12. R.-F. Kylix (Munich : from Arch. Zeit.) ... .

13. R.-F. Amphora (Munich : after Furtwangler)

14. From R.-F. Kylix (Rome : App. des rom. Inst.)

15. R.-F. Kylix (Berlin: after Juthner)

16. Panathenaic Amphora in B.M. (B 140)

Two Notes on Pylos and Sphaeteria.

1. General View Northwards from Cliffs South of the Panagia
Plan of Pylos and Sphaeteria (after Grundy) . . .

2. Red Bluff from South

3. Red Bluff and Route between Cliff and Bushes

4. Looking North from same point as preceding ...

5. Looking down from the Notch southward

6. Pylos and Sphaeteria from the North

7- » „ from the N. boundary of Voithio-Kilia

XI
PAGB

250
252
253
257
25 'J
2G0
261
261
262
263
264
264
265
266
268
272

275

•-':•;

277
278
279
280
282
283

PROCEEDINGS.

SESSION 1906-7.

The First General Meeting of the Society was held on November 13th,
1906, when the Rev. G. C. Richards read a paper on 'The Ionian Islands
in the Odyssey,' the object of which was to bring before the notice of
English students the theory of Prof. Dorpfeld that by Ithaca Homer in the
Odyssey meant the island later known as Leucadia or (after its chief town)
Leucas, and in modern times as Santa Maura. This theory is now
conveniently published in pamphlet form along with a reply to Prof, von
Wilamowitz (Athens, Beck & Barth). Since the excavation of the sixth
city at Hissarlik, the substantial accuracy of the descriptions of scenery in
the Iliad has been demonstrated, but the Odyssey has presented such
geographical difficulties as apparently to exclude personal knowledge
on the poet's part. The greatest difficulty is, however, removed by M.
Berard's identification of the Pylos of Nestor with Samikon, near the
mouth of the Alpheius, which, if correct, supplies an instance of the
transference of a place-name to another site. Dorpfeld's theory starts from
the comparison of Od. ix. 21 with xxi. 347, which shows that the three
islands Dulichium, Same and Zacynthus are off Elis, and Ithaca is not.
The only four islands worthy of being reckoned in the Septinsular
Republic (Corfu, Paxo, and Cerigo, not being in question) are Cefalonia,
Thiaki, Zante, and Santa Maura. The first three are off Elis ; Santa
Maura remains for the Homeric Ithaca. The ancients thought of Leucas
as an island, but as one that had been in earlier days connected with the
mainland : they therefore identified it with the peninsula in Od. xxiv. 57S,
and were debarred from identifying it with Dulichium or the Odyssean
Ithaca. Recent researches have shown conclusively that Leucas was an
island in 1000 B.C., and separated from the mainland then, as now, by a
channel liable to become c loked unless artificially kept open for naviga-
tion. This explains the transport of cattle from the mainland (Od. xiv.
IOO;, where the Cephallenians then lived (Od. xx. 187); and also the four
times repeated line 'I do not think you came by land,' which it is
impossible to interpret as a joke of Telemachus at the moment of recognition.
If Leucas = Ithaca, Cefalonia suits Dulichium well (Dulichium, if a real place
in the catalogue of Iliad A, cannot be imaginary in the Odyssey), Thiaki
is Same ; while Zante has always kept the same name. Thiaki will not
suit the Homeric data, (i) It is an island divided almost into halves,

xliv

with two mountains of approximately the same height, not an island with
one conspicuous mountain {Od. ix. 21). (2) It is not 'furthest of all to
the west.' (3) It is so close to Cefalonia that it seems to be part of it
from the eastern side (contrast with this ix. 25, xxi. 346). (4) Yet it
•yOayLcCkr) means low-lying, it is quite inappropriate to it ; whereas Strabo's
interpretation 'near to the mainland' suits Leucas, and if the other
rendering is correct, Leucas has more level land on the coast. (5) The
only possible site for the Megaron of Odysseus has yielded no trace of pre-
historic settlement to the excavations of Dorpfeld and Vollgraff. (6) There
is no possibility of identifying Asteris (Od. iv. 844) with the rock of
Daskalio. (7) The local identifications in the Thiaki are all modern and
suspicious ; the island was deserted, and only repeopled early in the
sixteenth century. Leucas provides (1) a suitable site for Odysseus's
home, where Dorpfeld has found prehistoric remains ; (2) similarly suitable
sites for the other Odyssean descriptions ; (3) a suitable Asteris with a
double harbour in Arkondi, between Santa Maura and Thiaki. Changes of
population (which Dorpfeld connects with the Dorian invasion) pushed the
Cephallenians into the islands (Od. xxiv and II. ii). The inhabitants of
the northern island passed over into Same and founded a new Ithaca
there ; while the inhabitants of Thiaki founded a city in Cephellenia,
which existed in historic times under the name Same or Samos. This
explains the statement of Pliny (H.N. iv. 15) that Neritis was an early
name of Leucas. It is impossible to maintain any longer that by Ithaca the
Odyssey means Thiaki. Against the view that the poet had no correct
local knowledge, and merely gave his fancy play (Von Wilamowitz), must
be set the ease with which Leucas satisfies the data of the Odyssey.

On November 27th a special general meeting was held for the purpose
of further discussing the paper read on November 13th. No one was
found to maintain the claims of Thiaki adequately to represent the Ithaca
of the Odyssey, as still maintained by Berard, and in Germany by those
who, like Menge, Michael, Lang, have opposed Dorpfeld's view. — Prof.
Ernest Gardner said he took up the position of a sceptic rather than of a
convinced opponent of Prof. Dorpfeld's theory or a defender of the identifica-
tion of Thiaki as Ithaca. Prof. Dorpfeld's arguments seemed to him to fall
into two classes : those which dealt with the geographical position of
the islands, as described or implied by Homer, and those which suggested
a minute topographical identification of sites, such as the stalactite cave
of the Nymphs or the double harbours on Asteris. The latter were
rather a source of weakness than of strength to the theory ; but it must be
admitted that the broader geographical evidence for Leucas made, in Prof.
Dorpfeld's masterly exposition, a very strong case, if we were to recognize
the Homeric topography in existing islands. We must, however, remember
that this theory would imply that the Odyssey was composed by a poet
and for an audience familiar with the Ionian Islands, and before 1000 B.C.,
from which time to the present day the names of the islands had been as
they now are. Such a solution of the Homeric question required a revision

xlv

of the whole evidence, philological, historical and literary, as well as
topographical, before it could be accepted ; and in any case the Odyssey
was interpreted by all the Greeks of the historical period as it is by modern
scholars. To them the Homeric topography did not correspond to any actual
topography ; and there did not, after all, seem sufficient reason for rejecting
the view now generally held that the poet's imagination rather than his
familiarity with the spot was responsible for his descriptions. Such a view
was more in accordance with the usual custom of poets and writers of
fiction. It was generally admitted that in the Odyssey we had an inner
zone, confined mainly to the Aegean, within which the geography was
familiar to the poet and his readers ; and an outer zone of vague traditions
and travellers' tales, where the knowledge of both was at best taken
at second hand. If we regarded the Ionian Islands as belonging to the
vague rather than the more definite region, there was no difficulty in keeping
to the accepted traditions about the names of the islands. — Prof. R. C.
Bosanquet said that minor identifications were of less importance, and
general correspondences alone should be looked for. On the whole,
Leucas reproduced Odyssean geography better than Thiaki. Dorpfeld's
finds in Leucas suggested to him an earlier d ite than the period generally
described as Mycenaean. The transference of names was extremely likely,
and had parallels in mediaeval and modern Greek history. But he was
not disposed to accept Dorpfeld's view that this took place at a very early
date. — After the reader of the paper had made a brief reply the President,
in summing up, regarded the claims of Thiaki as conclusively disproved,
but maintained that Homer could not be regarded as a safe source
for history.

The Third General meeting was held on February 19th. Professor P.
Gardner, President, was in the Chair and spoke as follows : —

Since our last meeting, one of the most distinguished of our Vice-
Presidents has been somewhat suddenly carried away by death, Professor
Henry Pelham, President of Trinity College, Oxford. He was from the
first a Member of the Council of this Society, and a Vice-President from
1895. In the foundation of what may be called offshoots of this Society,
the British Schools of Athens and Rome, he took an important part :
the latter was indeed a special child of his and he was Chairman of the
Committee of the School. Ever since our Society was. founded Professor
Pelham has been its earnest supporter at Oxford, and has done all in his
power to further its aims.

His work and his interests lay rather in the direction of Roman than of
Greek antiquity. But while an acknowledged master in his own studies,
he by no means limited his interest to them, but in a broad and earnest
spirit applied his great powers of organization and his strong personal
influence in support of the whole movement for broadening and deepening
classical study, for promoting research, travel and excavation, for spreading

xlvi

an interest in the inscriptions and the monuments of the ancient world, in
which this Society is so deeply interested. Though he never himself
contributed to our Journal, he did so copiously through his pupils.

I have often felt that if Professor Pelham had chosen a political career,
he would have attained a very high position. He had all the qualities of a
statesman. But he preferred the more modest career of a University
teacher and organizer. And his justification has been that his presence
and work at Oxford has raised the whole tone of the place. More I think
than any other man has he succeeded in imparting a high purpose to
Oxford study and a high tone to University business. All this was the
result of a noble personality. An English gentleman of the highest type,
straightforward, manly, open minded, ready to appreciate any kind of
excellence, generous almost to a fault, he was everywhere a central figure,
the doyen of ancient history at Oxford, the leader whom we were all glad
to follow. His departure leaves a great void which those who remain must
try between them to do something towards filling.

Mrs. S. Arthur Strong, LL.D., Litt.D., then read a paper by Professor
J. Strzygowski (printed in this volume, pp. 99-122). The paper was
discussed by Miss Gertrude Bell, Sir H. Howorth and Mr. Arthur Smith.

The Fourth General Meeting was held on April 30th, Mr. G. F. Hill in
the chair. Prof. Ridgeway read a paper on ' The True Scene of the
Second Act of the " Eumenides " of yEschylus/ of which the following is a
summary. His object was to inquire whether the true scene of the second
act was really the Erechtheum on the Acropolis, or whether we ought not
rather to look for another site. It would be said, What more appropriate
spot than on the Acropolis and at the most famous shrine of Athena in
the ' strong house of Erechtheus ' ? But the action required a shrine
which contained an ancient bretas, at which manslayers took sanctuary,
and moreover a bretas called by the name of Pallas, not of Athena ; for
the Pythian priestess speaks of Pallas ; Apollo bids Orestes take refuge
with Pallas, and it is Pallas who will see that he has a fair trial ; and the
Eumenides on their departure address the goddess as Pallas, though
Orestes twice, and the Chorus twice, speak of Athena. Now there is no
evidence that there was any such bretas in the Erechtheum or on the
Acropolis, or that such bretas ever conferred sanctuary ; whilst there is the
strongest evidence that the goddess of the Erechtheum was only known as
Athena, or the Polias, or Athena Polias, never as Pallas. It is still more
strange that not one of the four famous courts for the trial of homicide was
situated at the Erechtheum or on the Acropolis, though in the Prytaneum,
on the northern slope, were tried weapons which had shed the blood of
men or oxen. It seems incredible that iEschylus should not have placed
the trial at one of the four places where from of old manslayers were tried,
for the Attic audience would have been very censorious if he had placed
the trial at a spot where there was neither sanctuary nor law court. There

xlvii

were five courts for the trial of bloodshed : (i) the Areopagus, on the hill
west of the Acropolis, where were tried those accused of wilful murder,
poisoning and arson ; (2) the to eVt UaXXaSio) south-east of the Acropolis,
outside the walls, where were tried those guilty of involuntary homicide
(toU uKovo-L'tis airoKTeivciGi,) ; (3) the Delphinium, a shrine of the Delphian
Apollo, where those who pleaded justification (for instance, for having slain
an adulterer) were tried ; (4) the court at Phreattys, on a tongue of land
at Zea, where a man who was said to have shed blood during his period of
exile was tried, docked in a boat off the shore, the judges seated on the
land ; (5) the Prytaneum, already mentioned. It is obvious that the last two
cannot have been the scene of the trial in the play. The Areopagus will
not do, for there is not a jot of evidence for the existence of any ancient
image there called either Pallas or Athena, Pausanias mentioning only an
Athena Promachos ; nor is there the slightest evidence that there was
ever an asylum there. Again, the Delphinium will not do, for it certainly
did not contain a bretas of Athena, but rather an image of Apollo;
moreover, its name shows that it was not an immemorial cult-spot, since it
was in honour of the Delphian god, who first urged in Athens the plea that
deliberate homicide could be justified. Only the court of the Palladium
remains. Here there was a most ancient xoanon or bretas. This bretas
was an asylum, for each year the image was taken down to Phalerum to
the sea, doubtless to be washed in order to rid it of the pollution of the
manslayers who in the course of the year had embraced it, as Orestes is
supposed to have done (cf. Eur. Iph. Taur. 1169). The only name ever
applied to this image was Pallas or Palladium. Some said that it was the
Palladium from Troy ; others that Athena, after slaying her playmate
Pallas, in atonement set up an image of her. Finally, the court for trying
involuntary homicide in classical times was held there. (1) The plea urged
for Orestes is that he slew his mother on compulsion by Apollo, and Apollo
bears this out. (2) Apollo urges justification. It may be said that
justification trials were held at the Delphinium, not at the Palladium in
classical times ; but it has just been shown that the Delphinium is a later
court, as its name implies, and it derived its title from the story that
Apollo in the trial of Orestes had urged that certain kinds of homicide
could be justified. There is no evidence that the Delphinium was
ever an asylum. Hence we are led to conclude that in early days,
when the first step was taken towards mitigating the dread doctrine
SpdaavTt TraOelv, those who could plead that the)- had shed blood either
by mistake or justifiably took refuge at the Palladium. The trial of
Orestes is represented by /Eschylus as the first for murder: the court
which tries him is called a #607x6?, a term always applied to immemorial
institutions. The judges here, at the Delphinium, Phreattys, and
Prytaneum and in early times on the Areopagus, were the Ephetae, the Court
of the Fifty-one, i.e. 50 Ephetae and the King Archon. This court probably
was a survival of the ancient king and the Gcrousia, the only tribunal
in a primitive community. All the conditions required for the scene of

d 2

xhiii

Act II are now fulfilled : (i) an ancient image, (2) called Pallas, (3) used
as an asylum, (4) with a court attached for the trial of involuntary
bloodshed, and probably in early times for justifiable bloodshed also. But
not one of these conditions is fulfilled by the Erechtheum. It may be
urged that, though Orestes certainly took sanctuary at the Palladium,
nevertheless he was tried on the Areopagus ; but this involves the
insuperable difficulty that the man who had taken asylum would be
carried from that spot right away to another place, all the while being
exposed to the attacks of the avenger of blood. The essence of such
ancient asylums was that the case must be decided where the man was
in sanctuary. If Orestes took refuge at the Palladium, he must have
been tried at that court. Moreover he would be out of place in the
Areopagus, which tried cases of wilful murder only. — The paper was
briefly discussed by the Chairman and Prof. W. C. F. Anderson, the latter
expressing considerable doubt as to the proposed removal of the final
scene of the play from the Areopagus.

The Annual General Meeting was held at Burlington House on June
25th, the President, Professor Percy Gardner, taking the Chair. The Hon.
Secretary, Mr. George Macmillan, read the following report on behalf of
the Council : —

During the past session there has been no striking event to record, but
the Society has carried on its regular work in an efficient way and shown
abundant vitality in the several departments of its activity.

The modification in the rules recommended by the Council, that the
office of President be in future tenable for five years only, was approved
by members at the last Annual Meeting, and on the same occasion
Professor Percy Gardner was, under the terms of this rule, unanimously
elected President in place of the late Sir Richard Jebb.

The new departure in the Constitution of the Society, the creation of a
class to be admitted to certain privileges of the Society without payment of
entrance fee and to be known as " Student-Associates," was also approved
at the last Annual Meeting, but it is a little disappointing to find that
during the first year only three candidates have availed themselves of the
arrangement.

Professor Henry Jackson has been appointed a Member of the Editorial
Consultative Committee in the place of the late Sir Richard Jebb.

The Secretary, Mr. J. ff. Baker-Penoyre, has obtained leave of
absence for a year, which will be spent mainly in renewing or extending
his acquaintance with Greek lands, in seeing the latest results of excava-
tion and in independent research. Mr. Penoyre had earned some relief
after his strenuous labours for the Society and will no doubt come back
still better equipped for the varied duties of his post. The Council were

xlix

fortunate in securing the services of a member of the Society, Miss
Katherine Raleigh, to carry on the Secretary and Librarian's work in his
absence.

The continued interest which the Society takes in the progress of the
British Schools of Archaeology in Athens and Rome is emphasized by the
fact that a short abstract of the work of the two schools was inserted, by
special permission of the Council, in the volume of the Journal of Hellenic,
Studies for 1906. During the session of 1905-6 the efforts of the British
School at Athens had been rewarded by the discovery, on the site of
ancient Sparta, of the shrine of Artemis Orthia, the stern goddess in whose
honour Spartan youths underwent the ordeal of scourging. Thousands of
votive offerings were found there buried, among them a series of terra-
cotta masks which may have been used in some dramatic ritual. Early
in the present year another important discovery was made. The sanctuary
of Athena Chalkioikos on the Acropolis of Sparta was identified by
inscribed tiles found on the spot, and it is hoped that excavations there
may proceed next season. Among the finds on the site is a fifth-century
statuette in splendid preservation, representing a trumpeter. Further dis-
coveries of ivory figurines have since been made on the site of the temple
of Artemis Orthia. It is plain that the Society's grant of £100 for the
excavations in Laconia, the renewal of which was voted in January of
the present year, has been abundantly justified. The annual grant of £25
to the British School at Rome has been renewed for a further period of
three years.

The Roman School has undertaken, with the sanction of the Italian
Government, to make a new official catalogue of the sculpture in the
Capitol ine Museum. The work is well in hand and will shortly be
finished. Mr. A. M. Daniel was appointed Assistant Director of the School
at the opening of the session with the special duty of furthering this
enterprise.

The Library.

During the past year 277 visits to the Library are recorded, as against
375 for the year 1 904-5, and 372 for the year 1905-6, Besides those books
consulted in the Library 396 volumes were borrowed, the figures for the
preceding years being 312 and 415. 189 additions to the Library have
been made, including pamphlets, and exclusive of periodicals in progress.
The Council made the usual grant of .£75 for Library expense-.

Some interesting accessions are :— Hermann's Denkmaler der Malerei
dt-s Altertkums (in progress) ; Wiegand and Schrader, Priehe ; Wiegand,
Porosarchitectur der Akropolis ; the 5 Erganzungskefte of the fahrbueh

des k. k. d. arch. Instituts. The Library Catalogue published in 1903 has
now been brought up to the present date by adding the four supplements

under one cover. The price of the complete volume to members is 2s. 6d.
net, and that of the supplements 6d. net.

The Council desire to express their thanks to H. M. Government,
the Authorities of the University Presses at Oxford and Cambridge,
the Trustees of the British Museum, the Archaeological Institute of
America, the University of California, the Committee of the Archaeological
Society of Athens, the University of Athens, the University of Aberdeen,
the University of Colorado and the Institut National Genevois for
donations of books.

The following authors have presented copies of their works : — Dr.
Ashby, Mr. S. Chabert, Mr. J. W. Duff, Mr. S. Eitrem, Dr. J. W.
Evans, Dr. Farnell, Professor Fairclough, Mr. C. Gilliard, Mr. G. F. Hill,
Miss Hoste, Dr. Kenyon, Dr. Keser, Mr. G. Macdonald, Miss McDowall,
Mr. A. Malinin, Mr. F. H. Marshall, Mr. Phene Spiers, Mr. F. W. Simpson
Mr. J. W. White, Dr. A. Wilhelm.

Miscellaneous gifts of books have been received from Sir J. Evans, Miss
E. Fegan, Professor Ernest Gardner, Mr. F. W. Hasluck, Mr. G. F. Hill,
Mr. J. H. Hopkinson, Mr. Rawlings, the Rev. W. G. Rutherford, Mr.
Arthur Smith, and the Librarian.

The following publishers have presented recent works : — Messrs. E.
Arnold, Clark, Dent, Heinemann, Longmans and Green, Macmillan,
Methuen, Nutt, Reimer and Seernan.

The Collection of Negatives, Slides and Photographs.

During the past year the sale and hire of slides has proceeded briskly,
and many new negatives have been added to the collection. The
statistics will be published, as arranged, at the end of a three years'
period counting from 1906.

Members may find it convenient to know that there are in the
Library four complete copies of the Slide Catalogue (each with supple-
ments), and that these can be borrowed on the same conditions as the
other books.

The thanks of the Society are due to members of the Argonaut Camera
Club, members of the Hellenic Society and others, who have presented
lantern slides, negatives and photographs.

Finance.

It is satisfactory to be able to report that the Society's income for the
year has exceeded its expenditure by .£111. This surplus is less by £61
than that of last year, and a comparison with last year's accounts shows
this difference to be accounted for as follows : On the receipts side it will

be seen that the total income for the year is £87 less, the principal
differences appearing under the headings of Entrance Fees and Members'
Subscriptions in Arrear. The falling off in the receipts under the first
heading is explained by the fact that fewer new members have been
elected than in the year preceding ; and in the second case by the fact that
a number of resignations received have been those of members whose
subscriptions were in arrear and could not be recovered. On the expenses
side noticeable increase has to be reported only under the headings of
the Library— due to the completion of the catalogue— and the additional
.£100 granted towards the excavations in Laconia. On the other hand, a
saving has been effected on Sundry Printing, Postages and Miscellaneous
Expenses ; while it has not been thought necessary to write off any further
sum for depreciation of Stocks, so that the Treasurer is left with a balance
over on the year as stated above.

The account for the Journal shows that while the sales have dropped to
the normal average (the sales of back volumes in the year ending May
1906 were unusually high; the cost has also been less, the balance on this
account being almost identical with that of last year. The sale of five
copies of the Aristophanes Facsimile has well repaid the cost of a new
circular to Librarians, while the continued sale of the Supplementary
Volume on the Excavations at Phylakopi is also satisfactory. The Lantern
Slides and Photographs account shows this department to have again paid
its way, there being a small profit on the year.

Turning to the Balance Sheet, the surplus of Assets over Liabilities
shown is £283. The Debts Payable by the Society stand at the same
amount as last year, viz. :— £293, while the cash in hand amounts to £"613,
as against £376 last year, an increase of ^237. The Donations received
for the Endowment Fund during the year have amounted to £\6 16s.
The sum due for Arrears of Subscriptions at May 31st stands at £127.

The Council feel that the financial statement may be regarded as
satisfactory. It is hoped, however, that the Endowment Fund established
two years ago will not be lost sight of. The amount (£500) invested of
the sum already received has produced £\j in interest this year, and
the steady growth of this fund through Donations from members should
prove a very valuable source of future revenue.

Conclusion.

Hamdy Bey, the Director of the Museum at Constantinople, having
completed twenty-five years in that important office, the Council thought
it right, as he is one of our Honorary Members, to send him a congratu-
latory address in the name of the Society, and the compliment was
gratefully acknowledged.

In recording losses by death, special mention should be made of two

lii

Honorary Members, Professor Otto Benndorf of Vienna and Professor
F. Blass of Halle. Both were well known in *this country. Professor
Benndorf was always ready to encourage British scholars and explorers with
counsel and assistance, while Professor Blass had given much generous
and invaluable help to Drs. Grenfell and Hunt in the decipherment and
identification of the Greek literary papyri. In Professor Pelham, the
President of Trinity, the Society has lost one of its Vice-Presidents and
a man who had taken a keen interest in its work from the foundation. A
special tribute to his memory was paid by the President of the Society at
the first general meeting held after Professor Pelham's death.

During the year 29 new members and 3 Student Associates have been
elected. 38 have been lost by death or resignation. The number of
members at present on the list is 918, and there are in addition 184
subscribing libraries (an increase of 14 in the year) and 38 honorary
members.

It will be seen that the Society has during the past session well
maintained its position in its various fields of work. The only dis-
couraging symptom is that the number of new members elected falls short
of those lost by death or resignation, so that there is a slight diminution
in the total. It is not at present serious, but it is very important, in
view of the obligations undertaken, that the Society's revenue should be
rather increased than diminished, and the Council trust that all members will
do their best to bring in new candidates. It is on the other hand satisfactory
to note that there has been an increase of 14 in the number of subscribing
libraries. *

The Chairman then delivered the following address : —

A Society like ours is an organism with a continuous life. We have
lived long enough to form traditions, and we have been more successful
than most societies in giving birth to other societies and movements for
the advancement of science. So long as I have the honour to be President
I shall do what I can to cherish this common and continuous life. It is
my special duty to contribute towards it by an annual address, whereby we
mark the milestones of our course, see what we have done and what more
awaits us in the immediate future.

Every society which has a continuous life is anxious to keep up a
connexion with the past by a commemoration of those who are lost to it
by death. One of the most cherished institutions of Athens was the
veKvoria, the feast of all souls, when offerings were brought to the family
grave. We too have year by year to note who of our members have
passed away, and what they have bequeathed to us. Fortunately the list this
year is a short one. Of prominent members we have lost but two, Professor
Pelham and Mr. Shuckburgh. Our greatest loss is certainly that of our
Vice-President, Mr. Pelham. It is true that his interest was centred rather

liii

in the history of Rome than in that of Greece, and indeed in Roman
constitutional history. But with that breadth and generosity which were
the basis of his character, he extended his sympathy to research in all
parts of ancient life. He was most helpful in the founding of this Society,
and from the first every attempt of ours to widen and deepen Hellenic
studies found in him a friend and ally. I have, however, at a previous
meeting spoken more fully of our loss in Professor Pelham. I am glad to
say that a project is now being carried out to establish a memorial of him
in the form of a studentship at the British School of Rome in connexion
with the University of Oxford.

.Mr. Shuckburgh, one of our earliest members, was also rather
concerned with Roman than with Greek history. But his work in editing
the orator Lysias, and in publishing a translation of Polybius, bore on
Hellenic studies. He was an active member of the teaching staff at
Cambridge, and his personal character helped to make his work effective.

Of our foreign honorary members two have died, Professor Blass and
Professor Benndorf. Dr. Blass is known as an extremely able and
many-sided philologist. His works on Attic orators and the New
Testament writers are of great importance. He often visited England and
Ireland, and was almost one of us. In recent years he rendered invaluable
service to Messrs. Grenfell and Hunt in their publication of the papyri
which they have found in Egypt. With Dr. Benndorf I had much to
do. For many years he stood at the head of the study of Classical
Archaeology in Austria, first as Professor in the University of Vienna,
later as the head of the Austrian Archaeological Institute, which was
virtually his foundation, and of which he was the guiding spirit. That
Institute has done a mass of important work. Its Jahrcslieftc is among
the most important of archaeological periodicals. The researches of its
members in Lycia, and the now progressing excavations at Ephcsus have
produced results of great value. Dr. Benndorf was a prolific writer, and
all his work stands at a high level. It is the more satisfactory to
remember that when we elected him as honorary member, he wrote me a
letter expressing in the warmest terms his pleasure, and saying that he
regarded our Journal z,s second to none in the value of its contributions to
Hellenic antiquities.

While speaking of the works of our deceased members I should call
attention to a volume which has recently appeared of addresses and papers
by Sir Richard Jebb, appropriately edited by Mr. Butcher and Dr. Vefrall.
It may be long before there arises another scholar so accomplished and so

id' in his sympathies as Professor Jebb. I hope that our Society will
always cherish the traditions which he represented. In particular the
Romanes lecture delivered by him at Oxford is perfect, not merely in taste
and expression, but also in comprehension, in its realisation of what ancient
Greece can contribute to modern ways oi' thought and feeling and action,
how what is best in it may live again in our ideals, and tend to counteract
the many perverting and vulgarising influences of modern lite. 1 here are

liv

many scholars, but few who really deserve the name of Humanist, a name
which since the days of Erasmus has scarcely been better earned by any one
than by our late President.

Individuals come and go ; each builds his little part of the fabric of
knowledge, and hands on the task to successors. Let us turn from our own
losses to the more cheerful subject of the progress made in the year in
Hellenic studies, a progress the rate of which varies from year to year, but
which never ceases.

Of the activities of the Society during the year, you have heard from the
Report of the Council. I am glad to find that the particular part of that
work with which in the past I have been associated proceeds with energy
and success. The Journal of Hellenic Studies, of which twenty-six volumes
have now appeared, has kept up its reputation for thoroughness and
originality. The issue of last year contains excellent papers in most fields
of Hellenic study.

I think that since I ceased to be Editor of the Journal, it has covered a
somewhat wider field, there have been more papers primarily historical.
1 am quite prepared to rejoice at this. Greek life in all the variety of its
manifestations was one. Each branch of Hellenic study throws light on
other branches. The history of institutions, of literature, of philosophy and
of art is but one history after all ; and no man can properly understand one
side of Greek history who has not some knowledge of all.

Any complete account of the gains of the year is beyond the scope of
this slight sketch, and it is the less necessary that I should weary you with
a long catalogue of our successes, since there is now published every year,
by the Classical Association under the editorial care of Mr. Rouse, a brief
but complete summary of them, a most useful little volume called The
Years Work in Classical Studies. I cannot speak too highly of the
admirable labours of the group of scholars who thus bring together the facts
which so greatly interest members of this Society. Their publication leaves
me at liberty to select for comment any discoveries and any books which
seem to me of greater and more general interest.

I will begin with the prehistoric age, a field in which English scholars
have for a long time past taken a prominent place. Mr. Arthur Evans con-
tinues his work at Cnossus in Crete, work which has reconstructed a splendid
and hitherto unknown phase of early Anatolian culture. Unfortunately
bounds are set to Mr. Evans' inexhaustible energy and enterprise by the
smallness of funds available. But this year he tells me that his researches
have brought to light a complete new wing of the great palace at Cnossus,
which imperatively demands excavation ; and my knowledge of Mr. Evans
leads me to think that in one way or another he will succeed in carrying out
his purpose. Mr. Evans has also turned his attention towards more fully
working out the material already available. He has mapped out nine
successive periods of Minoan history, early, middle, and later ; and it is
being by degrees discovered that the prehistoric remains of the Cyclades,
and even of Italy, may be classified on lines parallel to those which can

be fixed in Crete. There is now set up at the Ashmolean Museum in
Oxford a very extensive arrangement of originals and facsimiles classified
according to period, giving the student such a conspectus of the products
of Minoan civilisation as can be seen nowhere else, unless indeed at Candia.

I would recommend a book recently published by a member of our
Council, Prof. R. Burrows' account of the recent researches and discoveries
in Crete. It seems to me an excellent piece of work.

But I feel that, interesting as are these peeps into a pre-Grcek
civilisation in Greek lands, this Society must always regard with still deeper
interest the literary and artistic works which belong to the historic Hellenes,
and which embody that spirit which has been one of the two or three great
formers of European civilisation.

Sparta, Syracuse, Miletus ! What associations cling to each of these
great names ! To the early history of each, recent excavation has brought
contributions. A series of vase-fragments, found by the German excavators
at Miletus, which reaches back to the Mycenaean age, proves how very
earh- was the foundation of that Ionian colony which was not only in
power and wealth but also in age the mother-city of Greek Asia. The
ground plans of many of the most important temples and buildings of
Miletus have been traced ; and the site is gradually giving up its secrets.
Remains of the ancient Sikel people, found at Syracuse, have given us the
touching point between the ancient native civilisation of Sicily and the new
culture brought in from Corinth by the Greek settlers. At Sparta the
excavations of the British School have brought to light first the site of the
shrine of Artemis Orthia, and then that of the bronze-lined temple of
Athena Chalcioecus, strewn with innumerable votive offerings in lead and
terracotta. But I must not dwell on Sparta, our Sparta, nor anticipate the
• accounts of discovery which will later in the year be laid before you at the
annual meeting of the British School of Athens.

In Rome a most valuable date has been recovered from our knowledge
of Greek vases. In one of the primitive graves laid bare by Signor Boni in
the Forum, there was found a small vase of the proto-Corinthian class.
This little vessel, purchased for a few pence by some early Roman, and
given by him to some deceased friend, has a value which cannot be
exaggerated for determining the stratification of the site.

The slowness with which the results of the excavations at Delphi are
published is a matter for much regret. A certain number of plates and
photographs have appeared, but for the explanation of them we still have
to trust to old volumes of the Bulletin de Correspondence hilldnique.
German archaeologists arc losing patience, and in recent numbers of the
Athenian Mittkeilungen Drs. Pomtow and Bulle have published searching
papers on the geography and the monuments of the sacred enclosure, in
which some of M. Homolle's views are called in question. These papers
will, I imagine, not hinder, but facilitate the French publication. But we
cannot help feeling that the results of the great excavations at Delphi will
not be set forth in an orderly way until they are no longer fresh.

Perhaps one reason for the delay may be that M. Homolle has been
recalled to France to occupy an important position, while the staff of the
French School of Athens is busy with the renewed excavations at Delos.
These promise, in their way, to be almost as important as those at Delphi :
we are recovering the whole plan of a Greek city of commerce, with its
wharves and store-houses, its spacious private houses, as well as its sacred
buildings. The inscriptions found at Delphi and Delos are of immense
extent and the greatest importance.

I may mention a few of the books of the year which throw light on
Hellenic studies — among these are Mr. Walters' Art of the Greeks,
Mr. Freeman's Schools of Hellas, Mr. Tucker's charming Life in Ancient
Athens, and Mr. Mahaffy's enlarged re- issue of his most genial and
delightful account of the Progress of Hellenism. These are books which
do not appeal only to the learned, but which bring the fruits of Greek
thought and idealism to bear upon the studies and the life of modern
times ; and surely there never was an age which needed the leaven of
Hellenic culture more than ours.

Also, since no line can be drawn between the art of Greece and that of
Rome, I may add Mrs. Strong's valuable manual of Roman Sculpture, which
may be considered the first attempt to set forth in order the chief
monuments of the great nation which so long dominated the world.

Greek literature will naturally and necessarily in the minds of English
students hold a more important place than Greek art. The principles
embodied in both are the same, but we are as a people more literary
than artistic. I will not on this occasion discuss at any length the
discoveries in the literary field. The discoveries which come nearest to us
are those made by Messrs. Grenfell and Hunt of papyri in the graves
of Egypt. There is, however, not much to say this year in regard to these.
The explorers have had a last season at Oxyrhynchus and made
considerable additions to the literary papyri found last year, which
contained, as you are aware, new Paeans of Pindar and fragments of a
fourth century historian. This is the end of the Oxyrhynchus excavation ;
and we must most heartily congratulate the self-sacrificing scholars who
have, with infinite pains and patience, added so greatly to our knowledge
of the earliest literary manuscripts of Greece and of the history of
Ptolemaic Egypt:

The researches of the past year have not brought to light any work
of art so important as the Charioteer of Delphi, or the Aeginetan marbles
of Furtwangler. Perhaps the most remarkable statues found are two
archaic figures from Samos, one seated and one standing, both male
and fully draped. They are of the same heavy Ionian style as the seated
figures from Branchidac in the British Museum. The seated figure is
a portrait, unfortunately headless, of Aeaces, father of Polycrates the
Tyrant. A dated monument of this kind is beyond value: and it enables
us to push back the beginnings of Ionian sculpture to a somewhat earlier
date than that formerly assigned to them. At the other end of the history

lvii

of Greek art, the researches of Danish archaeologists in Rhodes have allowed
us finally, it may be hoped, to fix the date of the Laocoon in the middle of
the first century B.C. We have thus secured two points marking the
beginning and the end of the splendid development of Greek sculpture.

Of recent books on sculpture perhaps the most useful to students,
though not the most learned, is Dr. von Mach's series of 500 photographic
plates, good enough for ordinary purposes, and published at the price
of a guinea. A learned work which will specially interest us is the
Catalogue of the Museum of Sparta by two members of our school, Messrs.
Tod and Wace. We may hope that next year this will be followed by
a catalogue undertaken by the British School of Rome, comprising the
celebrated sculpture of the Capitol Museum. Catalogues are not only
valuable to researchers, but their compilation is the best and most educative
work that can possibly be assigned to students.

The study of Greek vases has in the past been greatly hampered by
the fact that it has only been possible satisfactorily to pursue it in the
vase-rooms of one of the great museums of Europe. Old engravings
of vases, such as those published by Gerhard and Lenormant, were not
sufficiently accurate to be trustworthy. Twenty years ago Prof. Benndorf
of Vienna greatly facilitated the study by his issue of Vorlegebldtter for
use in archaeological instruction. The great series of plates now being
published by Furtwangler and Reichhold carries accuracy even further,
enabling us really to examine even questions of style without journeying
to the Museums of Europe. A like service to ancient mural paintings
is being performed by Dr. Hermann in his great series of reproductions
of Pompeian and other frescos. It is a pity that the cost of these
works places them out of the reach of ordinary persons ; but at all
events they may be consulted in libraries such as that of our Society.
Meanwhile Mr. Walters' new book on the History of Ancient Pottery has
provided for the first time an adequate handbook, to guide those who are
taking up the study of Greek vases.

Perhaps no side of Hellenic life has occupied more of the attention of
English scholars in recent years than Hellenic religion. By a sort of tacit
compact our two old Universities seem to have divided between them this
fascinating field. At Cambridge Professor Ridgeway, Dr. Frazer and Miss
Harrison have worked on the prehistoric and primitive e'ements which
survive in Hellenic religiun. Dr. Frazer, in his recent Adonis Attis and Osiris,
has also discussed the foreign elements which made their way into the
popular religion at the time of Greek decay. At Oxford, on the other
hand, the higher developments of Greek worship have attracted scholars.
Dr. Caird has written an admirable work on the theology of Greek
Philosophers ; Dr. Lewis Campbell has given an account of the religion
embodied in works of Greek literature ; Dr. Farnell has published a most
elaborate and learned work on the Culls of the Greek States, whereof two
volumes have appeared this year. I wonder whether there is any member
of this Society who has talent enough to bring together all these various

Iviii

sides of Hellenic religion, and give us a complete account of its main
features. It would be a fascinating task, and even if imperfectly accom-
plished would be a great help to the purposes of our Society.

Before concluding, I should like to turn for a few minutes from the past
to the future, to see what tasks now lie before us, in what directions we may
hope to extend the field of Hellenic studies. In this work we have now the
co-operation not only of the Schools of Athens and Rome, but also of the
Classical Association, whose energy and enterprise is infusing fresh life
into humane studies, especially in the northern Universities.

At the recent International meeting of Academies, the project for a
great Thesaurus of the Greek language was considered, and advanced some
steps toward actual accomplishment. At the meeting Professor Bywater
represented the British Academy ; but the enterprise has been especially
connected with the name of our late President. He was warmly in favour
of it ; and if it is finally carried out, it may be regarded as in a sense
a memorial of him. I am glad to say also that the very original and
thoroughgoing studies of Mr. Norman Gardiner on Greek Athletic Sports
are likely to take the form of a book, which will I am sure be epoch-making
in the subject with which it deals.

I take this opportunity of informing or reminding the Society that the
third International Congress of the History of Religions will be held at
Oxford in September, 1908. Many continental scholars will come to
England to take part in it ; and one may hope that the study of Hellenic
religion will be among those which will profit by the contact of mind
with mind.

It is a far cry from this learned Congress to the Olympic Gaipes. These
also are to be held next year in England. Their interest is no doubt
mainly practical. But it is worth while to pause and mark the influence
of Greece shown in the very fact that these international contests are called
Olympic. There still lingers about them something belonging to ancient
Greece. And it may be well to try to profit by the occasion by bringing
before English-speaking athletes what is really best in the athletic spirit of
ancient Greece, the dignity, the love of beauty, the manliness which marked
the earlier celebrations of the Olympic games, and to point out how in
later Greece the games were ruined by professionalism and over
specialisation.

Another good prospect is offered by the probable intention of the
Carnegie Institution at Washington to regard exploration and research
in the lands to the East of the Mediterranean as not outside its scope ; and
there is a prospect that some part of its munificent endowment may be
expended in researches in which we shall have an interest, if not a share,
in Asia Minor and Syria.

Our own duty is most closely connected with the excavation of Sparta.
Since the Greek Government has liberally made over to the British School
the site second to Athens in Greece in historic, if not in archaeologic, interest,
it behoves us to strain every nerve to find the men and the money necessary

lix

for the full carrying out of so important a task. If however the plan which
originated with Prof. Waldstein meets the success for which we must all
hope, it may be that even the excavation of Sparta will take second place
in comparison with that of Herculaneum. But very little of that incom-
parable site has as yet been touched. The extraordinary difficult)' and
expense involved in cutting through so many feet of hard deposit has
delayed the work. But we must remember that a single Roman villa at
Herculaneum, that called after the Pisos, has bestowed on us not only a large
number of papyrus rolls containing important documents, but also a series
of statues and busts in marble and bronze of incomparable extent and
beaut) r . These arc almost the only works of Greek art which have come
down to us, thanks to the preservative power of the soil, in almost perfect
condition, and every visitor to the museum at Naples must have felt his
breath taken away by the number and the beauty of these works of Greek
plastic art. It is within the mark to say that, if we leave out of account the
Hermes of Praxiteles and the Delphic charioteer, the remaining fruits of
the great excavations of Olympia and Delphi are from the point of view of
the modern lover of art surpassed by the contents of the Herculanean
Villa, May we but find one such more ; and even the dullest of scholars
and the driest of historians will feel what great help in the realisation of
the past is given us by the researches of the present.

After the President's address the Report of the Council was presented
to the meeting and adopted unanimously.

The officers and members of Council as nominated were then declared
unanimously elected or re-elected. Mr. Arthur Smith was elected as Vice-
President. Professor R. Burrows, Mr. R. M. Dawkins, Mr. C. C. Edgar,
Mr. H. Stuart-Jones and Dr. Rouse were elected to vacancies on the Council.

The proceedings were closed by a vote of thanks to the Auditors,
moved by Dr. Sandys and seconded by Professor Waldstein, who spoke
hopefully of the outlook for the proposed excavation of Herculaneum.

A comparison with the receipts and expenditure of the last ten years
is furnished by the following tables : —

ANALYSIS OF RECEIPTS FOR THE YEARS ENDING :—

Subscriptions. Current

Arrears

Life Compositions

Libraries

Entrance Fees

Dividends

Rent

Endowment Fund

"Excavations at Phylakopi,
sales

"Facsimile Codex Venetus,
sales ,

Lantern Slides Account

31 May,
1898.

31 May,

1899.

31 May,
1900.

31 May,
1901.

31 May,
1902.

31 May,
1903.

31 May,
1904.

31 May,
1905

31 May,
1906.

31 May,
1907.

613

598

634

636

628

646

672

709

789

753

205

3 2

126

118

122

l6 3

179

185

202

147

154

168

173

5°

IOO

133

103

43
...

42
...

3°

52*
93*

475
28*

8*
15*

17
21*

17*
3*

787

813

945

990

998

1,047

1,292

i,390

1,814

1239

* Receipts (less expenses).

ANALYSIS OF EXPENDITURE FOR THE YEARS ENDING : —

Rent

Insurance

Salaries

Library

Cost of Catalogue

Sundry Printing, Postage, and

Stationery, etc.

Printing and Postage, History
of Society

Printing and Postage, Pro-
ceedings at Anniversary

Lantern Slides Account

Photographs Account

Cost of Journal (less sales)

Cost of Journal, Reprint of
Vol. XXIII

Grants

" Facsimile of the Codex
Venetus of Aristophanes"...

'• Excavations at Phylakopi "...

Commission and Postage per

Bank

I (eprecial ion 1 if Stocks of
Publications

£
So

15
5o
93

5i6

i5
60
61

26
536

73
5S

390

74
61

i5
382

31 May, 31 May, 31 May, 31 May, 31 May, 31 May, 31 May, 31 May, 31 May 31 May,
1898. 1899. I 1900. I 1901. 1902. 1903. 1904. 1905. J 1905. 1907.

60
82

367

80
16
89
5o
55

72 137

125 150

948 960

35
454

250

210
156

} -

5ii

225

30
140

o 916 865 1,432 1,335 i,573 1,095 10 °9

165

100

147
24
10

5*
5"

122
260

13
176

158

100

14
178

556

125

356

225

123

104

Expenses (less sales).

J. H. S VOL. XXVII. (1907). PL.

J. H. S. VOL. XXVII. (1907). PL II.

J. H. S. VOL. XXVII. (1907). PL. III.

R.F. PELIKE IN THE BRITISH MUSEUM (E. 395).

R.F. FRAGMENT AT WURZBURG.

THROWING THE DISKOS.
[Plates I.-IIL]

A.— The Subs.

'Then the son of Peleus took an unwrought metal mass which anciently
the mighty Eetion was wont to hurl.' 1 This a6\os avroxoayvos, which was at
once the weight to be thrown and the prize in the Homeric competition, was
apparently a mass of pig iron just as it came from the furnace ; probably, as
Mr. J, L. Myres suggests to me, ' the contents of one of the old open-hearth
furnaces of the Mediterranean world, the natural unit quantity for the
purveyor and buyer of the metal, the classical analogy to which is the mass
of iron (p,v8po<; o-iStjpeos) which the Phocaeans threw into the sea before their
voyage westwards.' -

The word aoXos occurs only in this passage of the Iliad and in later
imitations of Homer. 3 The latter part of Iliad xxiii. is generally admitted to
be a somewhat late interpolation, and the interpolator, wishing to insert in
the games of Patroclus a competition similar to that of throwing the diskos,
gives to his interpolation an archaic appearance by the use of the unusual
word cr6\o<?. The diskos itself was sufficiently familiar in Homeric times for
the term 'a diskos throw' to be used as a measure of distance. 4 There is no
reason for supposing that the Homeric diskos differed essentially from that with
which we are familiar in later times, save that the earlier form was probably
of stone instead of metal. But in the aoXos avroxowvos we have the primi-
tive type from which all lifting and throwing competitions have arisen : a
stone, a mass of metal, or a tree trunk provides for early man a weapon in
time of war, a test of physical strength in time of peace.

Of these primitive contests and feats of strength we have various records
in Greece. A mass of red sandstone discovered at Olympia bears a very
early inscription to the effect that Bybon with one hand threw it over his
head. 5 The latter part of the inscription is unfortunately doubtful. Accord-
ing to Dittenberger and Purgold, it runs Bvficov Ti]Tepj] ^epl virepKe^aXd

1 Iliad xxiii. 826 ff.

'-' Il.lt. i. 1':.".. 4 //• xxiii. 431, 523 : cp. ii. 771:

:; Apollonius Rhodius iii. 1 :". 6 1 » , 1372; iv. 626; viii. 129, 186; xvii. 16S.

«;."i7, 851 : Nicander, Ther. 905 ; Nonnus, ehr. von 01 ; Jiithnei

Dioni/s. xxxvii. 667 ; Quint. Smyrnaeus iv. Turngerdthc, p

U.S. — VOL. XXVII. B

E. NORMAN GARDINER

pC inrepefidXero 6 <J>o[A.]a. Neither vTrepefidXero nor vTreptcefyaXd seems very
satisfactory, and Mr. H. B. Walters therefore suggests the reading virep-
K€(f>a\a.<; v-rrepefiaXe to o icpopet, ' he threw over his head this thing that he
was carrying.' But whatever the reading, the general sense is clear, and the
greater difficulty remains, how did he do it ? The stone weighs 143^ kilos,
and measures 68 x 33 X 38 cms. The explanation that lie merely lifted it
over his head 6 does not help matters ; a one-handed lift of an object of such
weight and shape is quite as incredible as the throw. I can only suggest that-
Bybon lifted it above his head with both hands, then balanced it on one hand
and threw it backwards. This is precisely what Aelian describes Titormus as
doing with a stone which Milo could hardly move. 7 First he raises it as far
his knees, then lifts it on to his shoulders, carries it eight opyvias, and throws-
it. A larger block of black volcanic rock weighing 480 kilos has been dis-
covered at Santorin. It bears the following inscription of about 500 B.C. :
EvfjbdaTas p.' dripev dirb %#oi>09 6 KpLToftovXov. 8 To lift such a weight off

the ground is a good performance, but quite pos-
sible. The only representation I know of such
scenes is in the interior of a r.-f. kylix in the
Louvre, G. 96, where we see a youth lifting in
both arms a large, roundish object, apparently a
stone (Fig. 1). Lifting feats are ascribed to Milo,
Euthymus, and other athletes, and the heroes in
the Iliad hurl in battle boulders that two stalwart
men can hardly lift. 1 '

The Homeric 0-0X09, like the stones described
above, has nothing distinctly athletic about it, any
more than our word ' weight.' It is merely a
mass of iron, and its athletic use is a mere acci-
dent. It is true that, like other objects, especially
metal objects or weapons, it has its history. It
belonged once to Eetion, king of Thebes, who
used to hurl it too. And Achilles after slaying Eetion brought it to Troy
among his possessions. But the author is far more concerned with its com-
mercial value: 'it will furnish a countryman with iron for five years'; and
we may feel sure that it found its place in Achilles's ship for practical'
purposes. In Apollonius Rhodius we shall find the 0-0X09 always mentioned
on the sea shore near the ships. A pig of iron would serve the sailor or
soldier in many ways.

The word must have been an unusual one, and certainly puzzled the
scholiasts, if we may judge from their numerous notes thereon. 10 With a single
exception to be discussed later, these notes, as well as the passages in later

Fig. 1.— R.-F. Kylix.
LouvitE, G. 96.

,; Chryssaphis Bulletin tin Comite" ii:s Jcux
Olympiqucs, 1906, p. 57.
1 Far. Hist. xii. 22.
8 1.0. xiii. No. 499.
: ' Aelian, op. cit. viii. 18; Paus. vi. 11 ;

11. v. 302; xii. 415. In the Odyssey the-
( \i lops and the Laestrygones hurl rocks at the
ships of Odysseus, ix. 481, 537 ; x. 121.

10 Jiithner, op. cit. pp. 20, 22, collects and
discusses the scholia.

THROWING THE DTSKOS. 3

authors where the word occurs, contaiu no information which may not be
derived from Homer. The word is said to have been found inscribed on a
bronze diskos discovered early in the last century in the bed of the Alpheius, 11
but without further information the evidence is worthless. We may however
probably connect the word with the names of various places, Soli in Cyprus,
and Cilicia, and Soloeis in Sicily (and Mauritania). According to Lewy, these
names are cognate to the Semitic sola, a rock, and Victor Bc'rard explains the
name as due in all these cases to a prominent headland or hill interrupting a
level stretch of coast. 12 The etymology is supported by the note of Hesy-
chius, ao\o<; = ovofxa fiowov ; and it certainly suits the Homeric croAo?, the
transition from a boulder to a mass of metal being easy. Thus in Apollonius
Rhodius the great round stone which Jason hurls into the midst of the
dragon brood is described as

Seivbv 'JLvvaXiov aoXov "Apeos (iii. 1366),

an expression which recalls the boulders thrown by the heroes of the Iliad in
war rather than an athletic implement.

But whatever the original meaning of the word, there is reason for think-
ing that it was from an early date appropriated to metal. Hesychius explains
aokoLTviros as fiv8po/cTV7ro<;, and the Soli in Cyprus and in Cilicia were certainly
in the neighbourhood of mines. So too Apollouius Rhodius in a remarkable
passage uses the word in connection with the Portus Argous, the modern
Porto Ferraio in Elba, the very name of which implies the neighbourhood of
the iron mines for which the island was already famous in classical times.
There the Argonauts on their voyage landed and

■^n]<f>2criv d7ro)fi6p^avTO Ka[x6vre<i
Ihpoi ci\i<; • XP 0L V ^ Kar ' alyiaXolo Keyyvrat
el'/ceXai • iv Be aoXoi kcl\ rev^ea decriceXa Keivcov (iv. 655).

Commentators and translators, following the scholiasts, identify the aoXos with
a sort of 6l<tko<;, and describe the Argonauts as holding athletic sports on the
shore of Aethalia, and then going through their ablutious with stones for strigils.
The pebbles on the shore have a flesh-like appearance in consequence, and their
discs and weapons are still to be seen there. 13 Did the Greeks never scrape off
the sweat except after games ? Had the Argonauts no need of such a process
after their endless wanderings and sufferings by sea? And when at last they
did find a resting-place, did they at once fall to throwing the diskos ? The
idea is preposterous, and but for the traditional explanation <t6Xo<; = &i<tko<; it
would never have been mooted. Why then are aoXoi mentioned '. Welther
is surely right, ' spectant fortasse ad ferri abundantiam.' 14 To Apollonius

11 CI.'!, i. 1541. As rain, ami tlie pebbles an' flecked as with

i- Lewy, Fremdicorter, p. 115 ; scarf-skin strigil-stripped

Berard, L'OdyssA et '<■ t, i. l>. 334. To this day ; ami their quoits and wondrous

13 Thus in the latest version by Mr. armour are there, all stone.

A. S. Way : 14 2>c Jy

There in athlete-strife 'lid they supple their geographicis, p. 96.

limbs till tin- sweat <>t them dripped

n -1

4 E. NORMAN GARDINER

Aethalia would at once suggest the pigs of iron exported from the island and
the mainland opposite ; and what objects could be more natural on board a ship,
whether they served for commerce, for ballast, for shipbuilding, or for weapons ?
Apollonins is surely thinking of the 0-0X09 avroxocovo^ in its commercial
rather than its athletic aspect.

However, as we have seen, the <to\o<; did take the place of the diskos in
the sports of the Iliad, and so the scholiasts interpret it. In the third and last
passage where Apollonius uses the word he describes Thetis coming to Aeaea,
where she finds the Argonauts beguiling the time,

croXft) pcTrjjcn r oLcrrodv (iv. 851),

T6p7T0/X€V0V<;.

The line is an elaborate, archaistic variation of the Homeric
hicncoiGLv repirovTO /cal alyavejjai teVre?. 10

The verse may seem to tell against my interpretation of the previous
passage, but in the case of a student and archaeologist like Apollonius it is not
unnatural that he should use the same word in two different senses, especially
when both senses can be justified from Homer. Even here the ao\o<; may
just as well be the weight as the diskos. Nicander, however, in his Thcriaca does
use the word for a diskos. It is with a crokos rebounding from a rock, he
says, that Apollo killed Hyacinthus. 16 Quintus Smyrnaeus and Nonnus
merely imitate and elaborate the passage Jn the Homeric sports, and their
evidence is worthless.

The scholiasts are much exercised in distinguishing the diskos and the
solos, and their artificial descriptions still find a place in our commentaries
and dictionaries. 17 The diskos is flat, the solos round and ball-shaped ; the
diskos of stone, the solos of metal : the diskos has a hole in it and a string
to throw it with, the solos is solid. The first distinction is fairly accurate ;
the diskos was flat, the solos a mass, more or less round. As to material, the
diskos we know was made both in stone and in metal, and probably the solos
could be either. The third and last distinction is ascribed to Tryphon, but
another version ascribed to Eratosthenes assigns the hole and string to the
solos, not to the diskos. 1S That they belonged to the solos is disproved by
every passage in which the word is used ; that they belonged to the diskos is
still more conclusively disproved by the monuments ; and Dr. Jiithner there-
fore rightly rejects the evidence of the scholiasts. But his idea that the
scholiasts invented the hole and string theory to explain the term irepLarpe^a^
is hardly satisfactory. The athletic craze had spread to Alexandria before the
time of Eratosthenes, so much so that few places produced such an array

15 //. ii. 774 = 0d. iv. 626; xvii. 168. Vdi/smj viii. 190 Schol. HHQT. 6 5<Wos

16 T/icr. 90. r <. \l6os -i)v. koX : Eparoadtvris iv 'OXvuTrioviKats

17 Jiithner, op, cit. pp. 19 ff., collects and luToptl rbv /uiv ooKov \iyaiv (Tidrjpovv J) ^vKivov
discusses the scholia at length. v) x a ^ K0 ^ v r€Tp7jfj.tvov Kara rb fxeoov iced tx o " Ta

18 Amnion. 40 SlffKOS juiv yi.p tort XlOos KaAiiSiov i^-r)!.Lixivnv, ou exoM-evoi fiaWovaiv ol
TfTpyfxtvos lis (prjfft Tpvcpoov iv Ttiix-mq irepl ayoovt£u/j.tioi.

EAArji'i(7 / uoC, coAoj be rh ^oA/coCi' 6\o(T(f>vptov.

THROWING THE DISKOS 5

of Olympic victors, 19 and we could bardly credit Eratosthenes with such
a blunder. It is possible, indeed, that he is speaking of Rome popular game
in which a round object was bowled along by means of a cOrd. 20 Mr. R. M.
Uawkins tells me he has seen such a game at Orvieto, round stones about a
foot in diameter being bowled along the sloping road, much to the danger of
pedestrians, by means of a strap wound round their periphery. Cheeses are
said to be thus employed in parts of Italy, and to be much improved by the
treatment. A more probable explanation, however, is suggested to me by
Mr. J. L. Myres. The scholia of Iliad xxiii., he says, have become
dislocated, and the string and hole belong not to the solos, but to the
Ka\avpo\jr mentioned a few lines further on. Polypoetes hurls the solos
' as far as an oxherd throws a fcaXavpoyp-.' This word, which is usually
explained somewhat pointlessly as 'a staff,' is really, says Mr. Myres, a sort of
bolas, a weapon consisting of a string with one or more stones attached to it,
which is used in Spanish America for throwing at and catching cattle. Mr.
Myres tells me that he has often seen Greek boys extemporising a sort of
bolas with a string and perforated stone. This explanation not only suits
the passage in Homer far better than the traditional one of the text, but also
offers a most satisfactory solution of the mistake about the solos and diskos.
From this passage the mistake would easily be copied elsewhere.

To sum up, aoXos is a heavy weight, originally perhaps a boulder,
afterwards a mass of metal, and in late writers it is occasionally used as
a synonym for diskos.

B.— The Diskos.

The word diskos means ' a thing thrown ' ; originally any stone of
convenient shape and size, then a stone artificially shaped for throwing, lastly
a similar object in metal. In Homer the diskos is still a stone : how far it is
artificial we cannot say, but in one passage at least it seems to be used of the
round, smooth stones that are found on the sea-shore. Odysseus, challenged
by the Phaeacians, picked up a diskos ' larger than the rest, a thick one. far
more massive than those wherewith the Phaeacians contended in casting.'- 1
The scene is the agora of the Phaeacians hard by the ships, and the sports
are of that impromptu, after-dinner sort that needs no apparatus. The
Phaeacians, as Alcinous admits, are no trained athletes, but ' swift of foot and
the best of seamen.' In the palaestra of the fifth century one would expect
to find diskoi of various weights, like the dumb-bells of a modern gymnasium,
bul surely not in the agora of the Phaeacians; yet Odysseus finds at one-' a
diskos such' as the Phaeacians themselves never use. If we think of the
diskos merely as a stone, the difficulty vanishes. The agora is hard by the
ships, and on the shore are diskoi ready to hand of all sizes, flat, smooth,
round pebbles such as fishermen use as weights for holding down their nets
and sails laid out to dry, and such as every visitor to the seaside inevitably

19 Krause Gymnaslilc, p. 800. •" Kietz Diakoswurf, p. -Jl - 1 Od. viii. 1S6 ft'.

E. NORMAN GARDINER

picks up to throw. From such a stone to the manufactured stone diskos the
transition is easy, and the recurrence of the phrase

Stcr/coHTiv Tepirovro kol aiyaveycn levres, 2 ' 2
together with the use of the term Sio-tcovpa as a measure of distance, makes
it probable that the manufactured diskos was known at all events in the
later Homeric times. It is sometimes stated, on the evidence of the epithet
KarcofjuaBioio, 23 that the Homeric heroes put the diskos as we put the weight.
It is possible, of course, but the evidence is insufficient; Karoo p,d8io<i would
be at least equally applicable to Myron's diskobolos, and the use of the terms
Sivrjaas and 7repLaTpe\jra(; proves that, whether they put the weight or not,
the Homeric heroes sometimes slung the weight.

In Pindar there can be no doubt that the diskos is an athletic imple-
ment ; and though he must have been familiar with the bronze diskos, he
makes his heroes Nikeus and Castor throw the older stone diskos. 24 The
latter is clearly represented on certain black figured vases as a thick, white
object 25 (PI. I.), but the evidence of the vases and of the actual diskoi which
we possess shows that the bronze diskos must have been introduced before
the beginning of the fifth century.

There exist in our museums various inscribed and carved marble diskoi. 20
But though in size and shape they differ little from the bronze specimens,
they are too fragile and thin for actual use, and their inscriptions prove
clearly that they are merely votive offerings. The practice of inscribing and
dedicating diskoi was an ancient one, as we may see from the diskos of
Iphitus dedicated at Olympia. With regard to the metal diskos we are more
fortunate. Of the specimens which we possess, four are probably votive
offerings, but one of these certainly, possibly three, had also been used ; the
rest were certainly intended for use. Most are of hammered bronze : four of
cast bronze, one of lead. Their weights and measurements can be best
seen from the following table : —

Finding place.

Museum.

Weight in
kilos.

Diameter

in cms.

Thickness
in mms.

1. Olympia

Olympia, Inv. 7567

5-707

5-13

2. Corfu

B. M. 2691

3-992

6-13

3. Gela

Vienna

3-800

4. Amyclae

Athens, De Ridder, Cat. 530

3-349

5. Olympia

Olympia, Inv. 4257

2-945 (?)

6-12

6. Olympia

Olympia, Inv. 12892

2-775

11-12

Rome, Museo Kireheriano

2-378

21, 21 -5

8. Olympia

Olympia, Inv. 2859

2-083

19, 22 5

3 at edge

9. Sicily

B.M. 248

2-075

10. Olympia

Berlin

2 023

17-5

9-10

11. Aegina

Berlin

1-984

12. Olympia

Berlin

1-721

13. Olympia

Berlin, Inv. 2286

1-353 (?)

20-5

14. Olympia

Olympia, Inv. 12891

1-268

4-12

15. Cephallenia

B.M. 3207

1-245

16-5

22 Ocl. iv. 626, xvii. 168 ; //. ii. 774.

23 //. xxiii. 431.

24 01. x. 72 wirpy ; Istlim. i. 23 \i0lvois S'ktkois.
23 B.M. Fuses, B. 134, 142, 271 ; Athens

Nat. Mas. 832 ; Munich 408 ; Berlin Vas. 1727.
28 Cambridge, Fitzwilliam Museum, 70, 72 ;
Kavvadias, VKvirra rod 'EQvikov Mover. 93 ;
Salzmann N6cro})ole de Camiros, PI. VIII.

THROWING THE DISKOS. 7

No. I 27 is of cast bronze, ornamented with concentric circles, and bear-
ing on one side a dedication by Poplius, a pentathlete of Corinth, to Olympian
.Zens, on the other the name of the alytarch, with the dates respectively
Ol. 255, 456. The difference in the date is possibly due to different methods
of reckoning the Olympiads. The style and weight make it improbable that
it was ever intended for actual use; if it was so, it is an illustration of the
degeneracy of athletics and the worship of brute strength which we see also
in the development of the heavy caestus.

No. 3 28 also of cast bronze had originally an inlaid dolphin, possibly of
silver, which would have added slightly to its weight.

No. 4 is of a somewhat scyphate shape.

No. 5 is fragmentary. The weight is calculated from the diameter and
thickness, assuming the surface to be spherical and the specific gravity thai ol
copper. If we make allowance for the slightly smaller specific gravity of
bronze and for the weathering of No. 6, the weights of the two will be
approximately equal.

No. 6. The weights of the Olympia diskoi are only approximate. They
were weighed, Mr. Bosanquet tells me, in the village shop.

No. 7 is of markedly scyphate shape.

No. 9 and No. II 29 are of cast bronze and engraved on the one side
with the figure of a jumper, on the other with that of a spear thrower. 50
No. 11 is also ornamented with a series of concentric circles. They belong
to the early part of the fifth century, but though they approximate closely in
weight and size to Nos. 8 and 10, their flatness and the sharpness of their
•edges make me doubtful whether they were intended for actual use.

No. 12 is of lead, and has probably lost considerably in weight.

No. 13 is imperfect. Three pieces are broken away from the edge. It
must have weighed 1*5 at least, perhaps considerably more. 31

No. 15 is also very badly worn, and must have been much heavier. It is
inscribed with two hexameters in archaic letters of the sixth century.

'E%cro/8a(9) fx! aveOrjKe Atfo? Qo(v)poiv fxeyaXoto
XaX/ceov a> vt/cacre Ke(f)aA(\)ava<; fieyadv/xuv^.' 1

In the last Olympic games a wooden diskos with a metal centre was
used. It is a thick clumsy object, the product of modern imagination,
utterly unlike and in every way inferior to the specimens which we have in
our museums. There is no authority for it whatsoever.

Is it possible from these data to arrive at any definite conclusions as to the
weights actually used in competitions ? The diskoi are all more or less worn,
and the weights are therefore only approximate. They seem, however, to fall

'-'" Olympia iv. 179 ; Jiithner op. cit. p. 28. concern us at present : I hope to ileal with it in

38 Jahreshefte ii. p. 201, PL I. another article.

- 9 Jiithner op. cit, pp. 27, "28. " For particulars oi these two 1 am indehted

3n The figure on the B.M. diskos is described to Dr. Zahn, from whom I received a di awing

in the Catalogue as an athlete holding a of No. 12.
measuring cord. The interpretation does not B.M. i 207.

8 E. NORMAN GARDINER

into certain groups. The best marked group is composed of Nos. 8 — 11, and
possibly 12 if Ave make allowance for the greater softness of lead. It suggests
a standard of 2*1. Heavier standards are suggested by Nos. 2 and 3, and by
Nos. 4 and 5, say 4*0 and 2*8, while Nos. 14, 15 point to a standard of 1-3.
Mr. Bosanquet, to whose kindness I am indebted for most of the data given
above, suggests that these different standards correspond to the different ages
of the competitors, but it is not safe to go beyond the general suggestion. We
know that in the treasury of the Sicyonians at Olympia three diskoi were kept
for the use of competitors in the pentathlon, and we know that there were two
classes of competitors, boys and men, and that the boys used a smaller diskos
than the men. 33 But we do not know that the standard at Olympia was the
same as that adopted elsewhere, or that the diskoi which we possess were
intended for competition. Rather we know that the metrical standards varied
considerably in different parts of the Greek world, and also that there
were different classifications of age at different festivals. 31 If then we are to-
arrive at any definite conclusion, we must confine ourselves to the Olympian
diskoi, and here the most that we can say is that Nos. 5 and 6 point to a
standard of about 2*8, Nos. 8 and 10 to one of about 21, which is confirmed
by Nos. 9 and 11 and perhaps 12.

Nor do the written records enable us to say for certain what was the
standard weight used by men. Phayllus is said to have thrown the diskos
ninety-five feet and Philostratus speaks of the hero Protesilaus throwing
beyond a hundred cubits, and that with a diskos twice the size of the
Olympian one. 35 Statins, 30 again, describes Phlegyas as hurling the diskos
across the Alpheius at its widest, 37 Little credit can be attached to these
records, but as far as they go they agree with the one fact emphasized by
writers, that the diskos was a heavy object. 38 In view of this and of the
existing diskoi, it seems probable that the men's diskos was usually con-
siderably heavier than 2 kilos ; usually, but not always, for the lightest diskos
in the list is that with which Exoidas defeated the high-souled Cephallenians.
The dimensions of the diskos in art correspond with those given in our
table. On the vases too the diskos is often ornamented with concentric
circles, as in Nos. 1 and 11, with various forms of crosses and dots, or with
the figures of birds or animals. 39 When not in use the diskos was kept
in a sort of sling, the two ends of which were tied in a knot. In such a
sling the diskos is frequently represented hanging on the wall or carried in
the hands of a youth. 40

33 Pausanias vi. 19, 3 ; i. 35, 3. free style 136 ft. The free style is possibly

34 e.g. at the Panathenaea, Nemea, and more effective, the modern Greek style certainly
Isthmia, 7raT5es, ayiveioi, &vdp«s. Elsewhere we less so than that employed by the ancient
have four or even five classes, C'.I.G. 1590, Greeks.

2214 ; l.G. ii. 444. 38 Lucian Anach. 27. Galen de red. luenda

35 Heroic, p. 291. ii. 9-11, iii. Homer, Statius loc. cit.

36 Thcb. vi. 675. 39 Juthner op. cit. p. 29.

37 In the last Olympic games a diskos of 40 ib. p. 30. E.g. B.M. Vases E 78 ; vol.
2 kilos was used. The winner in the Greek xxvi. of this journal, PI. xiii.

style threw it 115 ft. 4 in., the winner in the

Til ROW INC THE DISKOS.

G.—The Balbis.

The diskos, according to Philostratus, was thrown from a fiaXfik, our
knowledge of which is derived entirely from an extremely difficult passage
describing the picture of the death of Hyacinthus. 41 f3a\f3k BiaKex^pio-rat
/xiKpa /cal diroxpoxra ivl earom el firj to Karomv /cal to Begibv afceXos dve-
yovaa irpavi) ra efxirpoadev /cal Kovfyi^ovcra Odrepov rolv atce\oLi> o XP*I avv '
ava(3d\\e<r6at /cal av/xiropeveadai rr, Begin, to Be axv^ a T ™ ^aKov dvexovros
egaXkdgavra rr)v KefyaX^v eirl Beg id XPV icvpTOvadai Toaovrov, oaov viro&Xe
yfrai rd irXevpa /cal piirreiv olov dvtp,wvra /cal irpoo-ep,/3dXXovra toU BegioU
irdai. As Benndorf rightly says, ' omnia dependent a Myronis discobolo,'
the last sentence being a singularly happy description of the statue with the
ropelike pull of the right arm and the concentrated effort of all the right
side of the body. The first sentence is more difficult. ' A balbis is marked
off small and sufficient for one man standing el fiy to Karoinv, and support-
ing the right leg, the front part of the body leaning forward while it lightly
supports the other leg, which is to be swung forward and follow through with
the right hand.' This is the rendering given by Jiithner in an elaborate
discussion of this passage in Eranos Vindob., p. 317, and is manifestly superior
to Benndorf's version in his edition of the Imagines. Jiithner, after an
exhaustive examination of the -uses of fiaXfih, proceeds to identify the /3a\/3t?
of Philostratus with the stone slabs with which we are familiar from the
stadia of Olympia, Delphi, and other places. It is indeed highly probable,
though at present there is no proof thereof, that the pentathlete did throw the
diskos and the spear from the same line of slabs from which the races started.
But when Dr. Jiithner, assuming that the /9aX/3t? of Philostratus is identical
in size with these slabs, uses this assumption to explain the words el fir) to
/caroinv as due to the narrowness of the slabs, which only afforded room for
the right foot and not for the left foot behind, it is impossible to follow him ;
and, to do him justice, we must admit that he is not satisfied himself. 42 ' Ich
gestehe dass diese Erkliirung nicht ganz uberzeugend klingt.' As a matter
of fact, the starting slabs are amply sufficient for one man to stand on, though
possibly not in the position of Myron's diskobolos; but if we think out the
expression, what possible sense is there in ' a space small and sufficient for one
man except behind ' ? It is absolutely unthinkable, and we must either abandon
the passage as hopeless or find some other interpretation. The absurdity to
which we are reduced suggests the solution. A parallelogram cannot be
sufficient for one man ' except behind'; therefore the words qualify some-

41 Im. i. 24 (Benndorf and Schenkl). The authority, and, being the rarer word, is more

earlier text of iKayser reads SiaKix*^"" f° r likely to have been changed. The word is 30

5iok€x^P'°' to '> *J ^ for Kal rf, ffwavawaWtaOat wonderfully appropriate and lifelike,

for (Twava&oiXKtcTeai, and inserts ipyAfaat after u His reference of 5ia«€x^P" rTai t0 'ke

rit i^poadtv. Benndorfs text is undoubtedly two grooves on the starting slabs is quite

superior; but I regret the alteration of pointless.
(TuvavcnraMtodai. which has considerable MSS.

10 E. NORMAN GARDINER

thing else ; they can only qualify 8iatce)(a>pi<TTai. The meaning is clear at
once. The /3a\/3i9 is marked off — small and sufficient for one man — it is
marked off except behind. That the competitor must not overstep the line
in front is an elementary principle of all such competitions. In the present
day the hammer thrower or weight putter is confined to a square or circle.
The conditions for the diskos thrower were not so severe ; the balbis was
marked off sideways but not behind, and, as we shall see, the method of throw-
ing implied in the diskobolos of M/yron requires room for at least one step
forward. As I have said, it seems probable a priori that the starting slabs
should have been used for throwing the diskos and the spear. At the same
time, in the Delphi inscription describing the preparations for the Pythian
games we find mention of rap, irapa rots 7T€VTd0\oLs, the contract for which
was eight staters. 43 As special contracts are named for the running track, the
jumps, and the boxing ring (which would naturally serve also for wrestling
and the pankration), the arrangements for the pentathlon would seem to refer
to the diskos and the spear, i.e. the /3aA,/3/? and the means for measuring the
throw, and these arrangements seem therefore to be distinct from the starting
arrangements for the races.

Kietz's theory of the /3aX/3/? as a small platform sloping downwards has
been conclusively disproved by Jtithner, and need hardly have been noticed
had it not been -adopted by the Greek authorities in the recent Olympic games.
According to their wonderful regulations, the platform is 80 cm. long by 70 cm.
wide, with a height of not more than 15 cm. behind and not less than 5 cm. in
front. This extraordinary arrangement is based solely and entirely on the old,
•corrupt reading of the obscure passage in Philostratus quoted above. Even
if the old text was correct, its evidence would be absolutely worthless in face
of the manifest absurdity of the idea, and the fact that in all the numerous
statues, bronzes, vases, and gems representing the diskobolos there is not the
slightest trace of such a platform. Can we imagine Myron's diskobolos tilted
forwards ? Were it so, there would indeed be some excuse for Herbert
Spencer's criticism that he is about to fall on his face ! Even the scanty
literary evidence is conclusive against this arrangement. Lucian, Philo-
stratus, and Statius all emphasize the follow through of the diskobolos. As
the diskos swings down the left leg must inevitably be advanced, and a
platform which prevents such a movement not only renders a good throw
impossible, but, being fatal to all freedom and grace of action, is absolutely
un-Greek. 44

The throw was measured from the front line of the /3a\/^i9 to the place
where the diskos fell. That the competitor might not overstep the line in
throwing the diskos or the javelin is obvious, and in the case of the latter is
clearly implied by Pindar's expression //,?; repp,a 7rpo/3a<?, 4f ' words which could
never have been misinterpreted by anyone with even a superficial knowledge

43 B.C. II. 1899, p. 566, 1. 32. especially Mr. G. S. Robertson.

44 In my criticism of the modern Greek 4B Eustathius ad Horn. Od. viii. 202 p. 1591,
style I rely on the photographs in Chryssaphis's 42 ; Nem. vii. 70.

article and on descriptions from eye-witnesses,

THliO\V1N(.i TlIK DISKOS. 11

of athletics, except to defend some a 'priori theory. This line, like the
jumper's ftarrfp. is possibly indicated on certain vases by spears stuck in the
ground 40 (PI. II.). The place where the diskos fell was marked by a peg or
arrow, as described by Statius, and on several vases we see a diskobolos
putting down or pulling up such a mark. 17 (Fig. 2.)

In the modern free style the diskos is thrown from a circular area
2£ metres in diameter, and the method of throwing it is a modification of that
of throwing the hammer, the thrower's body making either two or three
complete turns. Of such a method there is no trace in ancient times, and.

Fig. •!.— K.-F. Kylikes. (After Jiithner, Fi

effective as it undoubtedly is, we may doubt if it would ever have been
invented but for the experience acquired in hammer-throwing.

D. — The Method of Throwing the Diskos. The Evidei

It would be tedious and unprofitable to describe and criticize the various
schemes elaborated by scholars for throwing the diskos. No branch of
Greek athletics has been treated at greater length or with less regard to
practice and unfortunately the scheme established by the Greek authorities
in the recent Olympic games is no exception. As the detects in these
schemes are largely due to a misappreciation of the value of the different
classes of evidence, it may be useful briefly to review the evidence.

(1) Literary.

The literary evidence is of the scantiest, and practically useless except
as confirming the evidence of the monuments. Besides the passage in
Philostratus discussed above, we have a few scattered allusions in Lucian and
a lengthy description in Statius of a type common in later epics. 48 In the
latter the heroic character of the contest is marked by the vast weight of the

4l! li.M. Vases B 574 ; Krause Qymn, xiv. Athenienne, f. 23 ; Pottier Louvre 9, 73.
A9 : bpi J.IT.S. xxiv. p. 186. u Lucian Philopseud. 18; Anachars. 27 ;

'" Jiithner op. cit. p. 32, Figs. 26, 27 : Coll. Dialog. Bear. xiv. 2 : Statius Theb. \
', Paris, L879, 79 ; Giranl U Education

12 E. NORMAN GARDINER

diskos. There are various archaeological details with which we are familiar
from other sources, but of the actual throw we learn nothing which we could
not learn from Myron's statue. The description of Hippomedon throwing
the diskos over his head into the air as a preliminary show off has indeed
suggested to M. Girard the delightful theory that the Greeks practised not
only 'le lancement en longueur' but 'le lancement en haut.' 49 One wonders
how they contrived to measure the height ! Jlithner, again, depends
chiefly on various poetical expressions for his wonderful theory of the
Kreisschwung. 50 a method of throwing the diskos by whirling the arm round
as when one jerks a cricket ball, a feat highly dangerous to performer and
spectators, but hardly likely to break records. He is surprised that a method
so frequently alluded to in literature finds such scanty support in the
monuments ! Lastly, it is on the strength of the passage in Philostratus and
of Myron's statue that the Greeks have derived rj 'EWrjvcK)) Sia/co/3o\ca,
arguing that because Myron's diskobolos has his right foot forward, the right
foot must be kept forward till the completion of the throw, and regard-
less of the fact that even the literary evidence proves that the left foot
was advanced as the diskos swung down. Such theories are highly creditable
to the imagination of the authors, and prove conclusively the inadequacy of
the literary evidence.

(2) Monumental.

Fortunately, the evidence of the monuments is exceptionally rich and
varied. The two statues, Myron's Diskobolos (Fig. 18) and the Standing
Diskobolos (Figs. 13), often assigned to Naucydes, are of first-rate importance,
such works being independent of the accidents which affect the types in the
lesser arts. Besides these we have a multitude of vases, bronzes, gems, and
coins representing this subject. Their evidence is of very different value.
Bronzes often form part of candelabra or serve as handles of vessels, and the
figure is therefore modified by practical considerations. The vase painter is
influenced by laws of composition or by the shape of the vase space, especially
in the interior of kylikes. The same cause operates still more strongly in
the case of gems and coins, as we may see by comparing the copies thereon of
Myron's statue with the original. 51 Hence, when we come to classifying the
types in these objects, we find apparent divergence, often due not to difference
in motive, but to differences of material or space, or to the age and style of
the artists. The classification is important, because the constant repetition of
any motive is fair evidence that the attitude represented is typical of the
performance. Again, not only does the style of the artist vary; that of the
diskobolos himself must have varied equally. It is inconceivable that the
ancient Greek athletes should have been compelled slavishly to imitate the
style of a particular performer, or even of a statue. The swing of the diskos
must have varied with individual performers as much as the swing of the golf

w op. cit. p. 202. 51 Furtwangler Antik. Ganmcn xliv. 26, 27.

•'" Op. cit. p. 32.

THROWING THE DISK' »s.

club, and we may naturally expect to see these differences of style reflected
in art. It has been necessary to dwell on these causes of divergence, because
of the tendency of archaeologists to force every attitude represented into one
series of movements. It is as though someone utterly ignorant of golf were
to try to reconstruct a single swing out of a miscellaneous collection of photo-
graphs or drawings of various golfers playing various shots and a few medals
or prizes bearing conventional representations of the game. At the same
time, though we must expect to find variety in style, \v<- shall find, I believe,
that the general principle of the throw is always the same.

E. — Typical Positions.

The scheme generally accepted in England till recently is based on the
two statues. It distinguishes three stages :

(1) The thrower takes up the position of the Standing Diskobolos, right
foot in front, the diskos in his left hand (Fig. 13).

(2) He swings the diskos forward and as it rises grasps it firmly with his
right hand, a position commoniy represented on vases (Fig. 3).

Fig. 3.— E.-F. Kvi.ix. B.M. E»?6

(3) He swings the diskos downwards and backwards in the right hand,
turning head and body to the right, till at the end of the backward swing he
is in the position represented by Myron (Fig. 18).

In this extremely simple scheme the right fool is the pivot on which the
whole body swings. This swing of the body round a fixed point is of the
nee of the swing of the diskos as of a golf club. The force comes not from
the arms, which merely connect the" body and the weight, but from the lift
of the thighs and swing of the body. Kietx - in his criticism of Six's
scheme " describes the upward swing in the left hand as useless, bi cause the

'■<''• IT- -

291.

H E. NORMAN GARDINER

diskos is thrown by the right hand, not the left. Such criticism shows
a complete misunderstanding of the whole theory of the swing, in which
the arms are less important than body and legs. The scheme as far as
it goes would be quite satisfactory, were it not that it fails to account for the
very large number of vase paintings where the diskobolos is shown with the left
foot advanced. The same criticism applies to Jiithner's scheme.

Besides the positions of the two statues, there are two other positions, of
such frequent occurrence on vases and on bronzes that we feel sure that they
belong to the ordinary method of throwing the diskos.

(1) The diskobolos .holds the diskos in front of him in both hands
(Figs. 3, 4, 5, 16, PI. II.).

(2) He hold the diskos flat in the right hand, which is turned out
so that the diskos rests on the forearm (Figs. 6, 7, 8, 9, 10, Pis. I., III.). The
left hand is usually raised above the head.

Let us examine these two types, paying especial attention to the position
of the feet, a detail of supreme importance, which has however been unduly
neglected by nearly all writers on the subject.

(1) Diskos held to the front in both hand*.

There can be no doubt as to the moment represented in this type : it is
the moment immediately before the diskos is swung back in the right hand.
If the right foot were always advanced, there would be no difficulty, this
position being the natural link between the positions of the two statues. In
a certain number of vases this is the case, 54 but in a far greater number we
find the left foot advanced (Fig. 4). Now the position of the feet cannot be
due to accident or carelessness, for the uniformity of other details is remark-
able. The advanced leg is always straight or nearly so, the other leg more or
less bent. The right hand always grasps the diskos, the left hand merely
supports it. This position of the hands is invariable. 55 Kietz, indeed,
mentions a kylix where the position of the hands is reversed ; but the figure
there depicted has nothing to do with the actual throw. 56 It is true he
holds the diskos to the front with both hands; but he is standing at ease, with
both feet together, and even has his himation thrown over his shoulder !
Seeing, then, that the artists show such care and accuracy in depicting the
position of the hands and other details, we cannot dismiss as accidental the fact
that in so many instances the left foot is advanced, while in both the statues the
right foot is in front. The difference cannot be that between a right-handed

54 Right foot in front, Gerh. A. V. 259, 260 position of feet. Schone Mus. Bocchi 1 (a frag-

(b.-f. vases) ; Mus. Chins, ii. t. 196 (r.-f. kylix) ; ment of kylix), r-f. krater, Naples 3084, r-f.

//..!/. Vases E 6 (r.-f. kylix). Left foot in kylix, Munich, 803 ; r-f. hydria, Munich, 377 ;

front. Arch. Zeit. 1878, 11, 1879, 4; Ann. d. I. r-f. amphora, Hermitage, St. Petersburg 1669 ;

1846, 4. M.; Gerh. A. V. 39, 294 ; Ant. Bild. i. r-f. fragment, Berlin 4041.

4, 68 ; d'Hancarville i. 68 ; B.M. l'ai,csB142, This very incomplete list will give some idea

326, 361, 576, 691, E 288 ; Mus. Greg. lii. 1 a, of the frequency of this type.

4, lviii. 1, lxx. 2 a, Bibliothbque Rationale, 354. 56 Mus. Chins ii. 195 ; Kietz op. cit. p. 75.

50 Cp. vases where I do not know the

THROWING THE DISKOS.

and left-handed throw, for, apart from the absence of any other evidence for
the use of the left hand, the position of the hands, the right hand firmly
grasping, the left merely supporting the diskos, proves conclusively that the
diskos is to be thrown with the right. We are forced to conclude, therefore,
that as the diskos swings forward in the left hand the left foot is advanced,
and of this we shall find further evidence when we come to consider this
forward swing.

How then does the diskobolos pass from this position with the left
foot forward to the position of Myron's statue ? The change of feet may
be effected in two ways — either by making another step forward with the right

Fig. ■!.— 15.-F. Kelebe. 15. M. E 361.

foot, or by drawing back the left foot. The former was the method adopted
by some performers in the Olympic games of 1S9G. Starting with the
left foot forward, the thrower raised the diskos in both hands to a level
with the shoulders, and at the moment of swinging it back advanced
the right foot, stepping forward again with the left foot as the diskos
swunc forward for the throw. This method requires room for three steps
forward, the impetus being helped by this forward movement The other
method requires room for only one step, and the pendulum-like swing of

E. NORMAN GARDINER

the left leg, first forward, then back, and finally forward again, seems at
least equally effective as helping the swing of the body, like the preliminary
waggle of a golf club. 57 Both methods are, of course, incompatible with
the balbis of the last Olympic games. Both are equally effective, and

possibly both were employed. A few of the vases
are in favour of the forward step, 58 but the strong
inclination of the body backwards in most of the
figures is in favour of the backward step. Par-
ticularly convincing is the attitude shown on a r.-f.
krater of Amasis (Fig. 5). The diskobolos is re-
presented three-quarter face, a position which has
greatly troubled the artist : he holds the diskos
before him in both hands, resting his weight on
his right foot, while the advanced left foot barely
touches the ground. The drawing is careless in
some respects, the athlete, for example, having
two right hands, but the balance of the body
clearly indicates that the left foot must be moved
backwards. A B.M. hydria B. 326, published in
Marquardt's Pentathlon, PI. IT., carries the move-
Fig. 5.— R.-F. Krater. ment a ste p f ar ther. The left hand is already
< tei ai wig, ig. . releasing the diskos, and the left foot is raised well

off the o-round. These two vases seem conclusive for the backward movement
of the left foot.

(2) Diskos fiat in the right haiid.

The second typical position of the diskobolos is with the diskos slightly
in front of the body in the right hand, which is turned outwards so
that the diskos rests flat against the forearm. The left arm is usually raised
above the head, or in a few cases it is stretched to the front. The right foot
is usually advanced. The attitude of the body varies greatly, from the stiff
upright position of certain bronzes to the stooping attitude depicted in the
interior of a r.-f. kylix ascribed to Euphronius (Fig. 9). This difference of
attitude seems at first sight to favour Kietz's view that we have here two
distinct types, one in which the body is practically at rest and the performer
is merely feeling the weight of the diskos by a short preliminary swing, the
other forming part of the actual swing. But a consideration of the various
classes of monuments leads rather to the conclusion that the position of the
arms is the essential point, and that the variation in the pose of the body is
due rather to the limitations of the early artists.

The type occurs in a number of bronzes, mostly archaic and of the class

57 This movement of the left leg is part of
the scheme prepared by Six, the only fault of
which is the endeavour to include too many

types in one series.

5 < E.g. B.M. Vanes E 393, 395.

THROWING THE DISKOS.

formerly described as Etruscan. 59 The diskobolos on the cover of a bronze
lebes in the B.M. (Fig. 6) is an imitation, possibly by an Etruscan artist, of

Fig. 6.— Diskobolos on Bronze Lebes. B.M. 559.

59 Reinach Mpertoire ii. 544, 3, 1. 5 ; Arch.
Am. 1904, p. 36, Fig. 8; Ann. d. I. 1879.
p. 133, Mo. 5 ( = Kietz, F'.g. 2) ; B.U. 1
502; Munich Antiquarium, L28 ; Bu
U.S. — VOL. XXVII.

F. A. Club 38 .. b. A

.similar type is found on the coins of A
Kietz, Fig. 4.

E. NORMAN GARDINER

the Greek archaic type, the general features of which it reproduces, though
it shows none of the promise of true archaic art. The stiff attitude with
the feet only slightly apart and the body upright, an attitude which at first
sight seems incompatible with vigorous action, is characteristic of early
bronzes. But what is the meaning of the uplifted left hand, unless it is to
balance the body ? And why should the body need this assistance unless
the diskos is being vigorously swung ? It seems as if the artists could
reproduce the position of the arms and legs, but not that of the body, in
action, a fact which should not surprise us when we remember that even
Myron in his diskobolos has not completely overcome this difficulty. This
view is confirmed by comparison with similar but finer Attic bronzes of the
fifth century, where the stiffness has disappeared and the attitude is full of
action and vigour. An excellent example of this is the beautiful little bronze

exhibited at the Burlington Fine Arts Club in 1903,
to the graceful vigour of which our illustration does
not do justice (Fig. 7). Here the right foot is well
advanced, the right knee is bent, and the weight,
as in Myron's statue, rests entirely on the right leg,
the left foot touching the ground only with the toes.
Another interesting bronze is figured in the Cata-
logue of the Forman Collection, No. 77. w The left
arm instead of being raised is extended horizontally
sideways, a variation very similar to one which we
shall find represented on the vases.

In these bronzes the right leg is usually ad-
vanced, but here, as in the vases, we find a few
exceptions with the other leg in front. 61 There are
also several bronzes representing a youth holding a
diskos in either right or left hand, but not expressing
any definite action. 1 '" 2

The stiff, upright type is found also on certain
black-figured vessels; for example, on the British
Museum Panathenaic vase B 143. 63 Kietz excludes
such vases from his consideration, on the ground that the figures are
depicted in a sort of procession. But despite the processional character
which is common to many early vases, the attitudes of all the athletes are
distinctive of their particular performances: the jumper swings his halteres,
the spear-thrower poises his spear, and the diskobolos, with his large white
diskos fiat in his right hand and his left hand uplifted, is identical in type
with the diskobolos of the bronzes. At the same time, the processional
character may account for the fact that he, like all the other figures in the

Fig. 7. — Fifth Century

Bronze.

(After Burlington Fine Arts

Club, 1903, PI. L.)

60 Reinach, op. cit. iii. 153, 6.

,;l ///. iii. 153, 5 (according to Reinach, identical
with ii. 544, 5, but this is apparently an error,
or else the drawings are wrong, as the position
of the legs is reversed) ; B.M. 504, 559.

6J Reinach, op. cit. ii. 544, 6, 7, 9 ; 545,
1, 2, 4 ; 814, 4.

63 Jiithner, op. cit., Fig. 44 ; J.H.S. vol. i.
PI. VIII.

THROWING THE DISKOS. 19

precession, has the left foot .advanced. The very similar figure on the B.M.
vase B 271 (PI. I.) has the right foot foremost.' 14 A very conventional
treatment of the same type is seen on a r.-f. amphora in the Lambert
Collection, reproduced by de Witte, PI. xxiv. The artist has attempted to
represent a diskobolos three-quarter face : the exaggerated treatment of the
muscles and the affected pose give the figure an archaistic appearance, and ue
Witte therefore considers the vase the work of an Italian imitator. I know of
no examples of the upright type on the red-figured vases, except perhaps a r.-f.
amphora in Munich, 65 on which are two groups of three figures, diskoboloi
and akontistai. Of these only the central figure in either group represent-
action. The diskobolos holds a diskos slightly to the front, flat in his right
hand ; his body is carefully balanced, leaning slightly backwards, with the
weight on the left foot, which is behind, but the left arm instead of being
raised is held by the side, bent at the elbow.

Just as we saw in the bronzes the archaic type, where the motive is
obscured by the stiffness, passing into a freer, more vigorous type where
the motive cannot possibly be mistaken, so it is with the
vases; and the connecting link is furnished by the disko-
bolos on a b.-f. tripod from Tanagra, now in Berlin 66
(Fig. 8). The artist's intention of expressing vigorous
action is obvious. The bearded athlete strides forward
with his right leg, holding a thick white diskos in his
right hand well in front of the body, the left hand being
raised as usual. In its stiffness and angularity the figure
resembles the archaic bronzes ; in its movement it suggests
the beautiful figure in the centre of the kylix in the "' t$ ERL[H 1707
Bourguignon collection at Naples, to which I have already
referred (Fig. 9). Every line and curve in this latter figure denotes action,
though I fancy the artist has somewhat exaggerated the stoop of the body to
suit the circular space.

The angle of the body naturally varies greatly; sometimes it is inclined
forward, sometimes upright, sometimes thrown well back. This latter
position is represented on a r.-f. kylix published by Noel des Vergers,
PI. xxxvii., and also on the outside of the Bourguignon kylix mentioned
above (Fig. 10). In the first of these the left foot is advanced ; in the second
the left arm instead of being raised above the head is stretched well to the
front, still, however, with the intention of balancing the body. We find
the same position of the arms on two r.-f. pelikai in the British Museum
E393, 395 (PI. III.). In both the diskobolos appears to be taking a step
forward with the right foot, but in the one the body is upright, in the other
it is stooping.

w So too has the diskobolos on a vase Munich, 408 ; Furtwangler - Reichhold,

figured by Tischbein i. "> I (Krause xiii. 14), of l'l. XLV*.

which I can fiud no particulars. Hi- lias his " ,; Berlin Yas. 17-7; Arch. Zeit. 1881,

righl ami bent at the elbow, ami the whole l'l. III.
attitude i- supple an 1 vigorous.

c -1

E. NORMAN GARDINER

If then we are right in assuming that, in spite of variations, the motive
of all these bronzes and vases is the same, the essential points being the use

F IG . g__R.-F. Kylix. Naples

PI. XVI.

of the unemployed hand to balance the body aod the position of the hand
which holds the diskos, it remains to consider what this motive is. We may

dismiss at once the delightful suggestion of Girard
that the diskobolos amused himself by throwing the
diskos up in the air and catching it, and that this
is the motive here represented. 07 We may also
dismiss Kietz's suggestion that in those cases where
a swing is clearly indicated the diskobolos is swing-
inp; the diskos backwards and forwards in the right
hand in order to make the muscles supple. This
idea fails to explain why the right hand is turned
outwards, and is based on the mistaken idea that
the throw of the diskos depends chiefly on the swing
of the right arm and not On the bod} 7 swing.
Juthner, again, imagines that the diskos is being
swung to the front in the right hand, and he there-
fore places this movement previous to the position
with the diskos in both hands. But this view is
open to the same objection as Kietz's scheme : there
is no danger of the diskos slipping, and the out-
ward turn of the right hand is pointless. It is not only pointless, it is
unnatural; for every gymnast knows how difficult it is to raise a weight

Fig. 10.

(From exterior of
Fig. 9.)

67 Lo-. cit. p. 202.

THROWING THE DTSKOS. 21

to the front with the hand turned out, and we may be sure that, in an
exercise depending for its success on the smooth and harmonious working

of all the muscles, any such constrained and awkward movement would
have been avoided. On the other hand, if the movement represented
is a downward swing from the two-handed position to the position of
Myron's statue, all difficulties vanish. This downward swing must have
been a very vigorous movement, in which the diskos might easily slip.
Hence the outward turn of the right hand to prevent slipping, and the
use of the left hand to preserve the balance. Hence too the fact that in
marly every case the right leg is advanced.

This view of the relation between the two types is confirmed by a
most interesting bronze in the British Museum (Fig. 11), which represents
the moment of transition. 68 The diskobolos stands with right foot
advanced and both hands holding the diskos, which instead of being
upright rests flat on the palm of the right hand, while the left hand only
touches it lightly and is on the point of letting go. We may notice, however,
that the thumb of the left hand is turned inwards, whereas in the vases it is
as a rule on the outside of the diskos. The same peculiarity is noticeable
in a bronze in the Museum at Athens, 09 where the diskos is held in both
hands high above the head. This position could not be reached if the diskos
were swung to the front in the left hand from the position of the standing
diskobolos. If, however, the diskobolos takes up his stand holding the diskos
in the left hand level with the shoulder, and then, grasping it with the right
hand, raises it to arm's length, we reach the exact position represented in the
bronzes. The attitude is indeed depicted on several vases, the diskos being
sometimes held close to the head, 70 sometimes extended to the front.' 1
Variations of the motive occur where the diskos is held thus on the right
hand ~- or rests on the shoulder, 7 ^ and in such figures there can be no connexion
with the actual throw, but the position of the hand in the bronzes forces us
to suppose a previous position with the diskos raised thus in the left hand.
The position is well shown on a lekythos from Eretria, published in the
Ephemeris for L886 (Fig. 12) though, as Mr. Bosanquet points out to me, it
i- doubtful if this particular vase represents a stage in the throw, the stool
with the clothes upon it rather indicating that the youth is leaving the

M B.M. Bronzcs675 ; Murray Greek Sculpture photograph of it. On a B.M. gem, 1817, the

i. p. 27-1. Perhaps the same moment is repre- diskobolos holds the diskos high above thi

seuted oil the B.M. kylix E 96, where the in both hands.

righl hand grasps the diskos as usual, while r " Gerh. A.V. 272; B.M. Vases, E 256.
the left, instead of supporting it underneath, <']>. relief on Athenian stele, Syhel Weft-
res fs flat on the surface. yeschichU der Kwist, p. 107; Murray i

"• No. 7412, Chryssaphis op. <■<(.. Fig. •_> Sculpture i. p. 183.

Though the surface is much corroded, it is of :1 R.-f. amphorae, Munich, 1, 9= Kietz

fine and vigorous workmanship. It will be op. cit., Figs. 6", 7: 'E<p. 'Apx- 1886, PI. IV:

Bhortly published in the catalogue of the B.M. Vases, E 96.

Athenian bronzes by the Ephor M. Stais, who Gerh. A.V. 22.

lias kindly sunt me through Mr. Dawkins a 7> A'..'/. Vases, B

E. NORMAN GARDINER

dressing-room, perhaps calling to a companion to come on. Here then we
have an alternative position to that of the Standing Diskobolos.

Fig. 11. — Bronze in the British Museum (675).

THROWING THE DI8KOS.

One difficulty remains. As we have seeu, the normal position of the
first type, with the diskos in both hands, shows the left foot forward ; the
normal position of the second type, with the diskos flat in the right hand,
shows the right foot forward. The transition is made either by advancing
the right foot or, more probably, by (hawing back the left. There are, however,
exceptions. Sometimes the right foot is to the front in the first type,
occasionally the left foot in the second type. 74 Though such variations may
possibly be due to the artist's carelessness, the care shown in other details
renders it more probable that they are due to variations in the style of
throwing. For example, a diskobolos starting with the right foot forward

i&?&PG fc&&tetef&< '^lara/s.

Fig. 12. — Lektthos from Eketria. (From 'E<p. 'Apx- 1886.)

might prefer to reach the forward position without advancing the left
foot. Or again, supposing he does advance the left foot, and supposing that
to reach the position of Myron's statue he has to draw back the left foot, this
movement of the foot may take place at various times. He may let go the
diskos with the left hand first, in which case we have the diskos swinging
hack in the right hand and the left les; still advanced. If, however, he drew
back the left leg first, he would for a moment still be holding the diskos in
both hands, but the right leg would be advanced, and it is noticeable that

74 V. supra pp. 14. IS.

24 E. NORMAN GARDINER

in vases which do show this attitude the left foot rests very lightly on the
ground, and the body is slightly inclined forward. The precise moment at
which the change took place would be just one of those details in which we
might expect to find a difference in style.

It has been necessary to discuss these types at length, because they
establish the two important principles, that the diskobolos changed the
position of his feet in different parts of the swing, and that there was con-
siderable variation in the style of throwing. At the same time, there are
certain typical positions which we may regard as fixed : the position with the
diskos in both hands, the swing back in the right hand, and Myron's dis-
kobolos. Bearing these principles and these positions in mind, we may proceed
to reconstruct the method of throwing.

F. — Reconstruction of the Throw.

(a) The stand and preliminary movements.

After first rubbing the diskos with sand 75 to secure a better grip, the
diskobolos takes up his position on the balbis, a space possibly marked out by
side line?, certainly by a line in front. At a little distance behind this line
he takes his stand, carefully measuring with his eye the space he requires, so
as not to overstep the line. This is the precise moment represented in the
Standing Diskobolos (Fig. 13), a statue the athletic meaning of which has been
so much neglected that one writer has actually proposed to reconstruct it as a
Hermes Diskophoros. 70 The care with which he is plantiug the right foot,
the firm grip which the toes are taking of the ground and the consequent
contraction of the muscles of the calf and leg, the slight bend of the body to
the right, all indicate that though the weight may for the moment rest on
the left leg, it will be immediately transferred to the right. Whether the
left leg is kept stationary, or is advanced to the front and then drawn back,
the right leg is the pivot on which the swing depends ; and on no other
hypothesis can the statue be explained. The position is one of rest ; but it
is the rest which precedes action, and every line of the figure betokens the
preparation for action. A point which has never, so far as I know, been duly
noticed is the direction of the head and eyes. The diskobolos is not, as is
sometimes asserted, looking down the course toward the mark, whatever the
mark may be ; much less is he taking aim, a part of the performance to
which Juthner and others assign an undue importance in a competition for
distance. His head is inclined to the right and somewhat downwards, and
his eyes are fixed on the ground a few feet in front. He is, as I have said,
mentally measuring the distance to which he may advance the left foot as in
the final swing the diskos is swung forward for the throw. For, in spite of the
modern Greek authorities, the actual throw must take place off the left foot.

75 Statins loc. cif. view is refuted by A. Michaelis in the same

7,; G. Habich in Jahrb. 1898. p. 57. This vol. p. 175.

THROWING THE DTSKOS. 25

The right forearm is said to be modern: if so, the restoration is peculiarly
happy : the position of the arm is found on certain bronzes which closely
resemble the statue 77 and the nervous curling of the fingers appropriately
suggests alertness and readiness to seize the diskos as it is swung forward in
the left hand to the front, position. The artisl has not merely put a diskos into

Fig, 13. -The Standing Diskobolos. Vatican.

the hand of a youth standing at ease; he has, as we should expect him to '\>>

in a work of such importance, selected a truly typical and important position.

Starting then in this position, the thrower may either keep the Left leg

stationary or step forward with it. In the latter ease he will be in the

Pourtales xiii. 3 = Reinacli ii. 545 Am. 1904, p. 36, n. 7.

E. NORMAN GARDINER

position shown on the Pauaetius kylix in Munich (Fig. 14). The left leg is
advanced and straight ; the weight rests on the right leg, which is bent ; the

body leans forward, and the right
hand extended to the front serves to
counterbalance the weight of the
diskos, which is still held behind the
body. Kietz sees in this figure a
left-handed diskobolos about to throw
— in a most original style! Juthner
sees in him a youth stooping down
to fix or take up a peg. But when
we compare this figure with others
which do undoubtedly represent this
motive, we find a fundamental dif-
ference in the whole attitude. The
' vorsichtig balancierender Schritt'
which Juthner himself notices is
surely not necessary for putting down
or pulling up a mark, and the
straightness of the extended left leg
would render such an-operation quite difficult. 78 Perhaps we may see the
swing forward in a more advanced state, the diskos being now in front

of the body, on a vase figured by Tischbein iv. 42 and on the B.M. kylix
E 58, 7i) but neither attitude is very satisfactory.

The position of the Vatican diskobolos is reproduced, as has been said, in

78 A very similar type occurs on the r.-f.
kylix in Rome, Mus. Greg. lxx. 2, a.

'» J.H.S. xxiv. 191, Fi«. 10.

THKOWTNCJ THE DISKOS.

certain bronzes, but does nut., so far as I know, occur od the vases. The latter,
however, suggest alternative methods of starting the swing. One of these,
where the diskos is held shoulder high in the left hand and then raised
above the head in both hands, has been already described. Another method
is suggested by vases which show a diskobolos holding the diskos in both
hands, but low down and with the arms bent close m to the body 80 (PJ. EL).
From this position it could be swung up in both hands to a level with the
head. In this type the left leg is already advanced. The actual swing
is perhaps depicted en the r.-f. kylix published in Gerhard A. V. 294
(Fig. 15), but it is possible that the position here represented forms part of the

Pig. lt.i. — iFi inieiiorof Fig. 14.)

Fig. 17.— R.-F. Kylix.
(After Hartwig, PI. LXIII. 2.)

first method described, coming between the moments represented in

Figs. 1+ and 16, and that in the other cases we have merely an athlete
carrying a diskos.

(b) 77/'' backward swing.

At the end of the swing forward, the diskobolos holds the diskos
extended to the front horizontally in both hands, the body upright
or inclined backwards, the weight chiefly on the back foot. If the
right foot is in front, no further change of foot is necessary; if the left
is in front, either the left foot mu*t be drawn back or the right foot
a Ivanced. The latter method does not explain the position with the diskos

R.-f. kylix in Louvre, figured by Pottier, viii. 1; Miu. Oreg. lii. 1, a; /•'..'/. V
Duris, F ; g. 6; Gozzadini di hi anl. .V.- B 576.

'ig. 18. — Myron's Diskobolos. (Rome, Palazzo Lanci
(By permission of Messrs. Methuen and Co.)

lotti )

THROWING THE DISK OS. 29

in the right hand and the left foot forward; and though there is some
evidence for it, most of the monuments are in favour of the drawing back ol
the left foot. The variations in the swing backward have already been
discussed. The diskos is held flat in the hand until it passes the body; at
this point the head and body commence to turn to the right, till at the
conclusion of the swing the position of Myron's diskobolos is reached. The
moment previous is well illustrated (Fig. 17 cf. below, p. 36) on a r.-f. kylix
figured by Hartwig. It is difficult to say for certain whether the moment
depicted is just before the top of the backward swing, or is the beginning
of the forward swung. But though the position of the feet certainly favours
the latter hypothesis, 81 the forward inclination of the body and the evident
care with which the youth is balancing himself seem to me conclusive ior
the former view, which is further confirmed by comparison with vases which
undoubtedly represent the forward swing.

An interesting variation of the top of the swing is represented on a
number of coins of Cos belonging to the early part of the fifth century. It
has been the fashion to connect this type with a totally different position
depicted on a Panathenaic amphora to be discussed later, and to place the
moment represented immediately before or after the top of the swing. A
few experiments would convince anyone that no one but a contortionist could
pass from this position to that of Myron's statue, or vice versa, and that such
a movement would be fatal to any success. The position of the right hand,
with the diskos turned to the front, excludes the theory of Chryssaphis that
we have here the beginning of the backward swing. Three points deserve
attention : the bend of the leg, the position of the right hand, and the position
of the left arm. When we examine a series of these coins we are led at once
to the conclusion that the attitude of the body is largely due to the shape
of the coin space. This will be obvious from the series published below
from the British Museum (Fig. 19). The way in which the body is bent over
to the right is manifestly impossible. In Myron's diskobolos the body turns
round the hips, but its inclination is forward ; here the turn is hardly indicated,
and the body is bent to the right. The explanation is, I believe, purely artistic.
The maker of the coin die wished to represent a diskobolos at the top of the
swing from the front. The difficulty of such a task can be best realised by a
glance at Myron's statue from this position. To the artist of the early fifth
century the difficulty was insuperable. The amount of foreshorten in-
required to represent the forward bend of the body was far beyond him, and
even if it had not been, the success of the result ona coin would have been more
than doubtful. Moreover, the circular space had to be appropriately filled.
He adopted therefore the obvious expedient of bending tin- body to the right
instead of forward. I am indebted to Mr. G. F. Hill for an admirable illus-
tration of the same process. In various cultus statues the arms were by the

81 The position of the right fool may be due to a dislike of foreshortening.

E. NORMAN GARDINER

side, bent at right angles, with the forearms extended to the front : on coins
representing them the forearms are extended not to the front, but to the right
and left. 82 Another illustration is afforded by the manner in which the heads
of the horses are turned right and left in a quadriga represented from the
front. 83 This view is confirmed by the variations which occur on the coins
themselves. The more the body is bent sideways, the more it is elongated,
while in the more upright figures there is a decided attempt at foreshortening.
To the same cause may be due the position of the front loot. The foot is
sometimes foreshortened in vase paintings, but the result is often by no means

Fig. 19. — Series of Coins of Cos in B.M. (Enlarged.

happy, and the coin maker therefore avoids the difficulty by extending the
foot in such a way that the diskobolos appears to be standing on tip-toe.

Coming to the arms and hands, we may remark first that the bending
of the right arm noticeable on certain of the coins is clearly due to consider-
ations of space. The position of the diskos, again, may be due to the fact

82 E.g. Artemis of Ephesus, B.M.C. Ionia Magnesia, ib. xix. 4, 5, 67.
xiii. 1, 2, 7, 8, VI; Artemis Leukophryene at 83 Gerh. ././'. 105.

THROWING THE DISKOS. 31

that if represented parallel to the body it would appear from the front
merely as a thin line, which on so small an object as a coin would be
unrecognisable. It may, however, also be due to a difference in the style
of throwing. We have seen that the left hand is sometimes raised above the
head in the swing back, and we shall find it still raised in the swing-
forward as represented in Fig. 21. It is only natural then that it should be
raised in the intermediate position. Now a supple, youthful athlete would be
always liable to exaggerate the swing, just as the youthful golfer does.
In such an exaggerated style the right hand would be raised higher than
in Myron's statue, and as it reached the perpendicular would naturally
turn outwards so that the diskos would face to the front, while the tendency
would be to keep the left hand raised in order to balance the body. I am
not saying that such a swing is as effective as that represented by Myron.
The artist of the coin was not depicting an ideal, but working from his
own experience of what may have been a local fashion. My point is
that such an exaggerated style is natural, and my point will be conceded
by any one familiar with the differences exhibited by golfers at the top
of the swing. Compare, for example, the position of the young St. Andrews
player with that of Vardon or Taylor. My conclusion then is that the
Coan coins represent a variation of the same moment as that of
Myron's statue, modified by the shape of the coin space and the
limitations of the artist.

(d) The throw.

' The diskobolos,' says Lucian, speaking of Myron's statue, ' seems as
if he would straighten himself up at the throw.' 84 At the besrinnino- of
the saving forward the extensor muscles come into play, and by a vigorous lift
from the right thigh the whole body is raised and straightened. This
momentary but most important movement is finely represented on two
vases, a Panathenaic vase in Naples and ab.-f. hydria in the British Museum
(Figs. 20 and 21). The attitude depicted is, as far as I know, unique in
Greek athletic art, which prefers positions of comparative rest and
equilibrium. But here we have a sort of snapshot, an impressionist picture
of a momentary position which cannot possibly be maintained. On the
Panathenaic vase especially, the thrower seems to be flying from the ground
in a way which recalls the figures of Winged Victory so strongly as to
suggest the idea that the attitude is borrowed from this type. The position
of body, legs, and arms is identical: substitute the victor's wreath for
the diskos, and add the wings, and we have the Winged Victory so often
represented on athletic vases; and it is certainly appropriate that the
artist should borrow from and suggest the figure of victory on a vase

w Pkilopseud. 18. Maiv tuv SiffKevovTa, ?)v 6i<\a£ovTa r& erf pep, toiKora avvavaarriaoui'vip
8 fyd,. (pTjs rov iirtKiKvcpora Kara rh a\V^ tt}s juera rfjj $oAijs ;
a<pt<jsu>s, airfffTpaujutfov eis rrjv 8i<TKO<p6pov, vpffia

E. NORMAN GARDINER

intended for the prize of victory. The position of the head and right
arm lead is similar to that depicted on the coins of Cos. But whereas on
the coins the body is bent sideways and the right leg is upright, on the

Fig. 20. — Panathekaic Vase. Naples. (After Bull. Nap.)

vase the whole figure forms a curve from head to toe, and is overbalanced.
On the British Museum hydria the curve is not quite as marked, and

Fig. 21.— R.-F. Hymua. B.M. E 164. (After li.C.H. 1899.)

the moment shown is slightly later ; the two vases illustrate also the two
different positions of the left arm which have already been noticed.

THROWING THE DISKOS. 33

A curious variation occurs in an early Athenian lekythos from the Pozzi
Collection, 85 to which Mr. G. F. Hill has called my attention (Fig. 22 >. The
general position closely resembles that which we have been discussing,
but the diskos instead of lying along the forearm is turned upwards so
as to rest between the fingers and thumb. 86 The vase painter frequently
makes mistakes in drawing hands, and such a mistake is the only possible
explanation of our present figure.

Jiithner, identifying the type of the Naples amphora with that
of the coins, considers them to represent a distinct method of throwing
the diskos, which he calls the Kreis-schwung. 87 He supposes the diskobolos
to whirl his arm round from the front, right over his head, and he supports
his theory by a variety of poetical quotations. Epeius hurls the solos
Sivijvas, Odysseus the diskos irepi<np£^a<; ; in Pindar Nikeus hurls the
stone x^P a tcvfcXaxrais. Even Propertius and Statius are called to witness.
It is hardly necessary to point out the obvious appropriateness of these ex-
pressions to the ordinary method of throwing the diskos or any other object.

Fi<;. 22. — Athenian Lekythos. Boulogne. (From Lc ilus6c.)

To argue that they denote a complete revolution cf the arm is the quintessence
of pedantry, inexcusable even if the expressions occurred in prose ; and it
is indeed surprising that Jiithner, whose useful work is generally dis-
tinguished by great soundness of judgment, should have allowed himself to be
misled by a passage or two of the scholiasts into so unpractical a theory.
A lio-ht object, easily grasped, might be thrown a certain distance in this way ;
certainlv not a heavy, slippery object like the diskos, much less the ponderous
Homeric solos. It is unnecessary to labour the point, especially as the
position which we are discussing forms the natural sequel to that of Myron's

diskobolos.

I have compared the position to that of the Winged Victory; but the
diskobolos has no wings, and unless he recovers his equilibrium by advancing
one foot, he must fall forward on the ground. The modern thrower ' in the
Hellenic style' does contrive to rid himself of the diskos in this attitude.
but the throw inevitably suffers; and there is absolutely no evidence that

« Lc Music, Vol. iii, p. 178, Fig. 12. !p. Figs. 11. 12. .•• eit. p. 32.

H.S. — VOL. XXVII. D

E. NORMAN GARDINER

the ancient diskobolos had to throw off the right foot. Indeed, the
evidence of literature and art is conclusive for the throw off the left
foot, the only rational method of throwing. 'The left foot' says Philo-
stratus, ' must be swung to the front with the right arm,' ss and his words are
confirmed by the less definite language of Lucian and Statins, and by the
vases. A r.-f. kylix at Boulogne 59 (Fig. 23) shows the early part of

Fig. 23. — R.-F. Kylix. Boulogne.

the movement. And the actual throw is vividly portrayed on a b.-f.
hydria in Vienna 90 (Fig. 24). On both vases the diskobolos strides vigorously
forward with his left leg. Elsewhere the motive is more or less obscured, but
Six appears to be right in thus explaining the figure of the diskobolos on the
Leyden Panathenaic amphora, though the grotesque exaggeration of this vase
discredits its evidence. 91 The diskobolos, the jumper, and the spear thrower all
appear to be running, an action perhaps introduced by the artist to give more
life to the convention of the processional type. If we make allowance for

this, we find the movements represented are really
typical of their respective performances. Possibly
we may assign to the same motive the diskobolos
on a r.-f. kylix of Corneto, published in the Mon.
d. I. XI. PI. 24 and also a wall painting repro-
duced in Mus. Chius. PI. cxxvi. In both thase
cases the bending of the right arm sugorests some
doubt as to the action, but this may be a modifi-
cation due to space limitation, such as we have
noticed in the representation of Myron's disko-
bolos on gems or on the coins of Cos. Of another
wall painting, figured in Mus. Borbonico, ix. 52, there can be no doubt.
Perhaps a stjll later moment, jusl before the diskos quits the right hand, is

Fig. 24.— B.-F. Hydria.
Vienna.

"' V. supra, p. 9.

H> Le Muste, Vol. iii, Fig. 32.

• Masner Mus. fiir Kun&t und Imdustn

Wien, 318.
U1 Arch. Zeit. 1881. PI. ix.

THROWING THE DISKOS. 35

represented in an exceedingly quainl tei racotta found a1 Smyrna and exhibited
al the Burlington Fine Arts Club in L888. 02 The exaggerated emaciation of the
body and contorted attitude are typical of this cl [ue. The position

is evidently influenced by Myron's statue, bu1 the left foot is advanced and the
diskos has already swung to the front.

The so-called bronze diskoboloi of Naples are said to represent the
moment after the throw, but this interpretation seems to me impossible, in
view of the position of the arms and the alertness and expectancy ex-
pressed both by the figures and the heads, and I have no doubt that they are
really wrestlers. Moreover, as the diskos leaves the hand the righl fool must
be again advanced to prevent the thrower from falling forward, and in the
bronzes the left foot is advanced. Of this final movement of the follow
through we have perhaps a representation in the right-hand athlete in Fig. 23,
but as the diskos has already left the hand, it is impossible with certainty to
identify the position. Whether in this movement the thrower was allowed
to overstep the balbis or not, we cannot say.

In palaestra scenes we often see youths carrying a diskos whose position,
whether at rest or in motion, has clearly nothing to do with the actual throw;
similar types occur also in the bronzes. Some of these have been alluded to,
and it would be useless to deal with them in detail. A word, however, may be
said as to certain gems, though the evidence of this class of monument has
little independent value. The numerous representatives of Myron's statu
have been already mentioned. Perhaps the influence of this statue may be
traced on a Berlin paste No. 4561, 94 where the diskobolos is apparently on the
point of throwing the diskos, which is swinging down in the right hand, still
behind the body, while the left arm is swung forward; the body and head are
thrown backward, and the right foot is vigorously advanced. Furtwangler
describes him as ' im Anlauf begriffen,' and compares him to the diskobolos
on the Leyden amphora. The evidence for a preliminary run is non-existent,
and I prefer to assign the position of the right leg to the influence of Myron.
Another Berlin gem shows a somewhat similar position of the arms, with the
left foot in advance. This gem is interesting from the fact that a flaw in the
stone was long interpreted as an elevation, or mound, representing the bater
or balbis, regardless of the fact that it cut off the thrower's left foot.
Fortunately we have a duplicate in which the flaw is wanting. 95 The dupli-
cate has also a peculiarity, in that the diskobolos holds in his left hand a cord,
a peculiarity repeated on the B.M. gem 1816. The cord is possibly a boxing
thong, but why if should be inserted has not been explained. The close
connection between the spear and the diskos renders it m »re likely that
it represents the spear-thrower's ayKvkri or amentum. It is curious that
such a piece >>\' evidence should have escaped the notice of the advoi
of the hole and string theory of the diskos or solos!

Bi mteghem Coll. 223.
vi Eurtwangler op. cit, xiiv. 26, -j7 . Ixvi. B ; . x!i\. -j- i =Krausi xiii. ! I

i:.M. ','■ 'wis, 742.

I. 2

36 E. NORMAN GARDINER

A summary of our conclusions may be useful.

1. The stance.

a. position of standing diskobolos. Fig. 13.
or b. diskos raised in left hand level with the shoulder. Fig. 12.
or c. diskos held in both hands level with waist. PI. II., Fig. 15.
From these positions, with or without a change of foot, the diskos is
swung or raised to

2. Position with left foot forward (usually) and diskos in both hands.

a. extended horizontally to the front. Figs. 3, 4, etc.
h. raised above the head. Fig. 11.

3. The diskos is swung downwards, resting on the right forearm. Either
before or in the course of the swing

a. the left foot is drawn back. Fig. 5.
or h. the right foot is advanced (PI. III.) so that we reach

4. The position of Myron's diskobolos. Fig. 18.

5. At the beginning of the swing forward the body is straightened.
Figs. 20, 21.

6. And as the diskos swings down, the left foot is vigorously advanced.
Figs. 22, 23.

7. Finally, after the diskos has left the hand, the right foot is again
advanced. Fig. 23.

I am again indebted to Mr. Cecil Smith for leave to publish objects in
the British Museum. To Messrs. J. L. Myres, R. C. Bosanquet, G. F. Hill,
H. B. Walters, and Dr. Zahn my obligations are many. The excellent illus-
tration of Myron's Diskobolos is reproduced by kind permission of Messrs.
Methuen from Mr. Walters's recently published ' Greek Art.' Mr. G. S. Robertson
has kindly given me the benefit of his experience both as diskobolos and as
Hellanodikes in the revived Olympic games, and it was a great satisfaction to
me to find that the conclusions which I had arrived at independently from the
study of the evidence agreed with those to which he had been led by practice.

The extremely interesting fragment of a Wiirzburg alabastron on
PI. III. is from a photograph obtained for me by Mr. Bosanquet from
Dr. Wolters, which arrived too late for notice in its proper place. The artist
lias depicted a back view of the position shown in Fig. 17. The legs
unfortunately are missing, the light patch visible below the arm being merely
a stain on the background. The drawing does not affect any of the views
put forward above, but no excuse is needed for the insertion of so original a
fragment.

E. Norman Gardiner.

THE PERSIAN FRIENDS OF HERODOTUS.

From what sources did Herodotus draw the materials for his history ?
At what date or dates did he compose it ? These inquiries have an endless
fascination for the student of Herodotus, which is not lessened by the fact
that they admit of no certain answer. The combinations which will be
suggested in this paper have, so far as I know, not been suggested before ;
but if, as is extremely likely, they have already been made, there is always a
certain interest in the fact that two inquirers, working independently, have
come to the same conclusions.

It is not necessary to give evidence of the fact that Herodotus himself
was highly satisfied with his own sources of information as to Persian history,
and that he considered he could speak with authority upon it. (Cf. e.g. i. 95.)
Nor is it necessary to give evidence for the view that Herodotus had on some
points official or semi-official Persian information : e.g. in his account
of the satrapies in Bk. III., of the Royal Road in Bk. V., and of the Persian
army in Bk. VII.

These two points will be assumed, and also that Herodotus is a trust-
worthy witness, that he reports truly what he has heard, without exaggeration
or suppression, and that he had some idea of the differing value of various
witnesses. The problem then is to find a Persian source from which Herodotus
could derive :

(1) Information that seemed to him trustworthy as to the rise of the
Achaemenid house, and its establishment on the throne.

(2) Official details as to the resources and organization of the Persian
Empire in the fifth century.

(3) Definite information as to the inner court circle of Susa. The story,
e.g. as to Amestris and the wife of Masistes (ix. 108 sq.) is told by Herodotus
with as much fulness of detail and with as complete a confidence, as the story
of the Philaidae in the Chersonese or that of Alexander of Macedon.

It is not suggested that these stories and others like them are to be
accepted by us as accurate, but only that Herodotus considered he had
full grounds for relating in detail events and motives which would be
unknown to ordinary informants, outside of court circles.

Now it can hardly be supposed that Herodotus, when himself in the
East, ever penetrated into the government offices, much less up the back-stairs
of the court. Even apart from his ignorance of all languages but Greek, he

38 J. WELLS

was only in the position of an ordinary traveller, seeing the wonders of the
great king's realm on sufferance. No Persian grandee, still less one of the
intimate court circle, would have unbosomed himself confidentially to an obscure
Greek, travelling in the company of merchants, and not improbably engaged
in business on his own account.

It may be maintained that Herodotus' informants were his own country-
men, who were either treading as exiles the antechambers of Susa or engaged
there professionally, as was Democedes, or Apollonides (Ctesias, 29, 42), 1 the
immoral physician from Cos. This seems, however, less likely, having regard
to two points :

(1) The accuracy of Herodotus' information as to Persian names, and the
fulness of his details on many matters which would be quite outside of the
sphere of interest of an ordinary Greek. The information we get from Ctesias,
the Greek court physician of the next generation, does not give us a high idea
of the sources of information open to, or of the accuracy of, the Greek
hangers-on of the Great King.

(2) Herodotus' own tone is always that of one who speaks with
authority, and who considers he has sure sources of information. Of course
this second argument will be worthless to those who look on Herodotus as an
inquirer prepared to accept any information, and prepared also to maintain
it was the best information, simply because he had it.

The assumption that Herodotus had real and special sources of informa-
tion as to Persian affairs, and the still more probable assumption that he did
not find these when himself in the East, lead us to the conclusion that
Herodotus must have met nearer home persons qualified to give him accurate
and detailed information on Oriental matters, under circumstances which
permitted him to question them carefully : such a source of information it is
usually supposed that he found in Demaratus (cf. Matzat, Hermes vi. p. 479
seq., and others), who may well have furnished Herodotus with many of his
details as to Xerxes' invasion. The object of this paper is to suggest another
and even more important source for his inner history of the Persian
Court.

The passage in Herodotus is of considerable importance ; he ends Bk. III.
{c. 160) with the words ' the son of this Megabyzus was Zopyrus, who went
over to Athens as a deserter from the Persians.'

The date of this desertion, and its significance will be considered later ;
first it is necessary to emphasize the fact that Zopyrus, if Herodotus really
met him, is exactly the informant who satisfies the conditions of our inquiry,
for he was one who was certainly able to give Herodotus the information
desired, and one moreover who was likely to give it just in the form in which
Herodotus reproduces it.

1 The references to Ctesias are ^iven to C. inaccessible; Valueless an t tie works of Ctesias

sedition, published with the history of are, a critical edition in a cheap and handy

Herodotu i Paris. F. Didot, 1844). This seems form would be of greal convenience to students

the edition must generally used, but it is very of Graeco-Oriental history.

THE PERSIAN FRIENDS OF HERODoTI's 39

The reasons for holding this are obvious:

(1) Zopyrus belonged to the inner circle of the Persian Court. He was
tlic grandson of Amestris, the terrible wife of Xerxes, and the nephew of that
monarch. Hence he would have known intimately the whole dark history
of court intrigues, and his story as told us by Ctesias (especially 2!». 42 — 3)
corresponds exactly to the picture of cruelty and lust on which Herodotus just
lifts the curtain.

(2) He was the son of Megabyzus, one of Xerxes' six generals in chief
against Greece (vii. 82 — 121). Hence he was in a position to know the
full details of the Persian army list, which Herodotus gives us at such
length in Bk. VII. Moreover this connexion would give him the detailed
knowledge of the stages of the Royal Road which Herodotus reproduces
from some Persian source in Bk. V. (cc. 52 — 3).

It may be added that the arrogant suggestion of an attack on Susa,
which accompanies the account of the Royal Road (v. 49), is quite in keeping
with the character of a Persian prince whose Hellenic sympathies have led him
to desert his country. It is of course quite out of place in the mouth of the
Ionian Aristagoras, who wanted only defence against the Great King.

(3) His grandfather had been governor of Babylon, and of the resources
of this satrapy Herodotus had especially full information (i. 192, iii. 92); it
must be added, however, that Herodotus gives these as they were under the
satraps that succeeded Zopyrus.

It will be seen then that Zopyrus had special facilities for giving official
information on two of the points (i.e. the Army and the Royal Road) where
Herodotus preserves it, and that on the third point, the organization of the
Empire, he had also some special qualifications for giving information, though
not to so marked an extent as in the two previous cases. When we turn from
Herodotus' information as to the present resources of Persia to his accounts
of its past history, Zopyrus again fits in with the requirements of our inquiry.
Herodotus of course had far too much information as to Persian history to
have derived it exclusively from any one source. But on two important
episodes at least Zopyrus was a particularly qualified witness.

(1) Herodotus' account of the conspiracy against the Pseudo-Smerdis is
in marked contrast to that of Ctesias in the accuracy of its name-, and
(perhaps it may be added) in the general correctness of its outline.

Now the grandfather and the namesake of the deserting Zopyrus had
been one of the Seven Conspirators, and the story of that crisis in Persian
history must have been a tradition in his family, and Herodotus may well
have heard it from him. This supposition throws considerable light on one
of the most disputed passages in Herodotus. If we assume that the historian
obtained from Zopyrus the famous account of the debate of the Seven as to
possible forms of government, we have at once an explanation of the' curious
and surprising insistency with which the historian maintains the accuracy of
his version (iii. 80, vi. 43), and also of its very un-Oriental character. Modern
critics rightly agree with the sceptics of Herodotus' own day in doubting the
authenticity of the speeches said then to have been delivered. Full of

40 J. WELLS

interest as these speeches are, they are interesting as giving us Greek political
ideas of the fifth century, and not as reproducing the sentiments of Persian
grandees of the sixth century. But the colouring is not that of Herodotus
himself: it is clearly derived from some informant, whom he considers of
special value. If we attribute the whole version to an occidentalized
Persian, who was yet the grandson of one of the conspirators, we have a full
and sufficient explanation at once of Herodotus' mistaken confidence and of
the curiously misplaced colouring which has offended critics from Herodotus'
own day to our own." 2

Again if we suppose that Zopyrus was Herodotus' informant as to the
conspiracy, we get a reasonable explanation of the serious blunder with which
Herodotus concludes his story. The historian is ignorant of the real claim of
Darius to the throne, and makes his winning it the result of a trick (iii. 84).
This perversion is exactly what we should expect from a Persian whose father
and himself had alike suffered at the hands of the Achaemenid family.
If anything is clear as to the inner history of Persia at this time, it is that
certain noble houses resented the predominance of one royal family, and that
Megabyzus was conspicuous for this independence. I must return to this
point later, but we may notice the same colouring in the remark with which
Herodotus introduces his story of Cyrus : ' Following the report of some of the
Persians, those" I mean who do not desire to glorify the history of Cyrus, but
to speak that which is in fact true ' (i. 95). The story that follows corre-
sponds to this introduction : Herodotus ignores the royal descent of Cyrus
from Achaemenes, although in Bk. VII. 11 he has rightly recorded the
names of the Achaemenid family. Herodotus' informant knew the facts as
to Darius' accession, but did not choose to draw the attention of the Greek
historian to them.

The other episode of Persian history which here especially concerns
us is the story of the second capture of Babylon in Bk. Ill (cc. 153 seq.).
It will be obvious to anyone that this account as a whole might well have
been derived from the grandson of the man who is the hero of the story, and
there are certain points in it which look like a special family tradition,
e.g. the details as to the mule prodigy in c. 153, and as to the special honours
to Zopyrus in c. 160 — ' no one of the Persians surpassed him (i.e. Zopyrus) in
good service, either of those who came after or of those who had gone before,
excepting Cyrus alone.'

Of the historic value of the story, I shall speak at the end of this paper.
So far I have tried to show that Zopyrus the deserter is exactly the informant

- I submit that this explanation of the well Protagoras. His theory has not a scrap of

known difficulty as to these speeches is tar more evidence in its favour, and E. Meyer (Forsch. i.

satisfactory than the view that sees in them 201-2) well says ' Maass makes Herodotus a

an instance of the composite character of simpleton, if he imagines that he could

Herodotus' work. Maass, e.g. {Hermes xxii. 581 impose on the public as historical facts inven-

seq.), on the strength of a supposed parallel tions of his good friend Protagoras.' Moreover

in Isocrates, argues that the historian has here the theory ignores Herodotus' insistence on his

introduced some of the 'negative arguments' own accuracy, which is surely a most important

(KcnafidAAoi>Tts Ac/701) of his contemporary point.

THE PERSIAN FRIENDS OF HERODOTUS 41

from whom Herodotus might have derived important passages in his work,
and that certain features in the narrative are more easily explained, if we
suppose he did so derive them, than on any other supposition.

There is one more passage in Herodotus which may well have come
from Zopyrus, i.e. the account of the unsuccessful attempt of the Persian
Sataspes to circumnavigate Africa (iv. 43). This account presents just the
same features as some of those which have been already considered, Le. there
is an intimate knowledge of the relationships of the inner court circle of
Susa, and of the cruelty and lust which prevailed there. The offence of
Sataspes was committed against the daughter of Zopyrus the deserter, and
may well have been one of the causes which inflamed his hatred and
jealousy of the Achaemenidae. Some suggest, however, that Herodotus'
source here is revealed in his concluding words : he describes how the
servant of Sataspes after his master's death escaped to Samos, and there was
robbed by a Samian whose name Herodotus knew, though he considerately
suppresses it. This part of the story must have been heard by Herodotus
in Samos, but it is not unlikely that he adds it as confirming from an
independent Samian source what he had himself learned from one who was
in the most intimate way concerned in the story.

But it is now r necessary to consider if Herodotus was likely to have met
the younger Zopyrus or indeed could have met him.

To answer this question we must consider the date of the Persian's
desertion. All our information as to this is derived from Ctesias (29.
33-43). Now that author seems, speaking generally, about the most
untrustworthy of our ancient authorities, and in his account of the events
that now concern us, he is clearly wrong on some points, e.g. he contradicts
Thucydides as to the name of the place where Inarus and the Greeks in
Egypt offered their last resistance to the Persians : he calls it Byblus (29. 34),
Thuc. (i. 109) calls it Prosopitis.

But it is obvious that Ctesias had means of knowing the inner history of
the Persian court, however badly he used those means at times ; he was
physician there in the generation after the events he is describing ; and, as he
had this department of his subject mainly to himself, he was not liable to
be misled in his details as to court-scandals by the burning desire to con-
tradict Herodotus which was so misleading to him in his account of more
important events. And his narrative as to Megabyzus and Zopyrus is
consistent in its main outlines with what we know elsewhere, and is confirmed
in one important point by an undesigned coincidence with Herodotus. Ctesias
makes Megabyzus die at the age of 76 (29. 41) ; this advanced age agrees
with Herodotus' account of that veteran, and especially with the detail (iv.
43) that he had a granddaughter of marriageable age before 465 B.C.

Assuming then, as is generally done, that Ctesias may be depended on
for these personal details, we have the following data for determining the
chronology of the family of Zopyrus. Megabyzus reduced Egypt, and
received the submission on terms of Inarus and the Greeks in 454, probably
early in the summer of that year (so Busolt, iii. p. 330). The vengeance of

42 J. WELLS

Amestris was delayed-for five years, but in the end the safe conduct was violated,
Inarus was impaled and the Greek prisoners were executed. This must have
happened then about 450. Megabyzus, angry at the violation of the terms
arranged by him, proceeded to revolt in his satrapy of Syria, and fought two
campaigns against his royal master. It may well be that the renewed attack
of the Greeks on Cyprus under Cimon (spring of 449) was connected with
this civil war in the Persian Empire, and that the reconciliation of the
rebel satrap with Artaxerxes, which followed in the year 448, was a part of
the same negotiations which led to the agreement (whether definitely
concluded or simply tacitly understood) loosely called the 'Peace of Callias,'
or the ' Peace of Cimon.'

It is surely permissible to conjecture that the Greek victories had their
natural effect at the Persian court and led to division of opinion there ; one
section of its grandees would urge that Persian policy should be modified,
and that the victorious Occidentals should be conciliated ; another section
would be confirmed by disaster in the old national traditions. If such a
division took place, Megabyzus was clearly the head of the Hellenizing
party in Persia : this is seen in his conciliatory attitude in Egypt, and agrees
with the story of Ctesias (in itself improbable) that he had declined to attack
Delphi when ordered by Xerxes (20. 27). The dc facto suspension of hostilities
between Athens and the Great King marked the triumph of the policy of
Megabyzus ; but so far as he personally was concerned, the Great King was not
disposed to overlook his independent spirit, and the too successful general,
having once more offended Artaxerxes, by interference in his hunting, was
banished for five years (29. 40-41). This banishment may be conjectured
to have taken place before the end of 448, in which case the final restoration
of Megabyzus to favour would fall about 443.

Ctesias gives no hint how soon his death followed, but goes on to tell of
the misconduct of his widow Amytis and her lingering illness and death.
We can only guess at the length of time required for these events, which
were immediately followed by the desertion of Zopyrus, but they can easily
be fitted into three years, and the desertion of Zopyrus will then fall in 440. This
year is probable in itself, for it is obvious from Thucydides' (i. 115) account of
the Samian revolt that the war party at the Persian court had the upper
hand in that year. That there was a connexion between the desertion of
Zopyrus and the general relations of Athens and Persia is not generally
recognized ; but it is probable in itself, and it is confirmed by the parallel
events of the next generation, when, if we may trust Andocides (de Pace 29 ;
cf. Busolt, III. 1354, 1417), hostilities with the Great King were precipitated
by the Athenian alliance with the rebel Amorges in Caria (cf. Th. viii. 5).
Perhaps the relation may be one of cause, and not of effect as has been
suggested above, and the desertion of Zopyrus may have led to the intrigues
of Pissuthnes (Th. i. 115) against Athens, not been caused by them. In this
case we should have to antedate the desertion by a year, i.e. place it about
4 H . The point cannot be settled, but either date, 441 or 440, can be fitted in
with the narrative of Ctesias.

THE PERSIAN FRIENDS OF HERODOTUS 43

The sequel of that narrative confirms materially the political im-
portance which has been assigned above to the conduct of Zopyrus. He
went, we are told (Ctes. 29. 43), with the Athenians against Gauuus in Caria,
and there met his death when endeavouring to negociate the surrender of
that town to the Athenians. This expedition most probably was connected
with the (roubles caused by the Samian revolt; Pericles (i. 116) himself
made a demonstration in the direction of Caunus in 440, and we know from
the tribute lists thai there was something like a general revolt in tin- ' < 'avian
quarter' of the Athenian Empire at this period (Busolt, iii. 554). So tar as
concerns Zopyrus and Caunus, we know (if we may trust Ctesias" that Caunus
remained for a short time under the authority of the Great King, for Amestris
was able to impale the unlucky Caunian whose hands had cast the deadly stone
against her traitorous grandson. But Caunus was again under Athenian
authority in 436 {LG. I. -44;, when it figures at the head of the list of the
' Ionian Tribute ' payers. Hence the death of Zopyrus must certainly fall
before this year. Perhaps we may suggest that the cruelty of Amestris worked
for Athens more effectually than the arms of Zopyrus : it was not likely to
stimulate loyalty to the Great King, when his subjects were impaled for too
successful a resistance to a traitor because that traitor was of royal blood. It
seems therefore that we may date the death of Zopyrus with fair confidence at
the end of 440 or early in 439. It must come in before the reduction of Samos,
and the restitution of the status-quo with Persia. Pericles, then at the
height of his influence, was not likely after this to provoke Persia by reckless
expeditions against Caria (cf. Busolt, iii. 544-5).

To sum up then this part of the argument. The desertion of Zopyrus
was not a mere personal freak : it was the act of a Persian prince whose family
had shown Hellenic prejudices before, and was connected with political events
of great importance: it probably took place in 441 or 440, and his death
followed within a year.

Before discussing the bearing of these dates on the life of Herodotus.
I must first refer to two other (and varying) dates which have been assigned
for the desertion of Zopyrus. Rawlinson (ad loc.) says : ' this is probably the
latest event mentioned by Herodotus. It is mentioned by Ctesias almost
immediately before the death of Artaxerxes, and so belongs most likely to the
year 426 or 425.' The 'and so ' begs the whole question : there is no causal
connexion between what Ctesias says of Zopyrus and what he says of
Artaxerxes. And it is most difficult to fit an Athenian expedition against
Caria into the years 426 and 425. And moreover had Herodotus kuown of the
death of Zopyrus, he would almost certainly have mentioned it : and it seems
that he must have known, had it happened after his return to Athens about
430 : this point, however, will be dealt with later.

Kirchhoff refers incidentally to the desertion of Zopyrus in his famous
paper 'Die Entstehungszeit des Herodotischen Geschiehtwerks ' {Abh. d.
K. A. d<r W. Berlin, 1878, p. 16), and calculates it. from the data given by
Ctesias, as falling between 445 and 431 (which is obviously true), but much
' nearer the latter date than the former;' this latter statement is. I think I

44 J. WELLS

have shown, quite unproven. Kirchhoff uses the point simply to prove that
Herodotus wrote the end of Bk. III. at a later period than the first two and
a half books ; the desertion, he argues, is one of the events of which Herodotus
was not aware when he went to Thurii, and of which he heard on his return
to Athens about 432. But Kirchhoff' quite fails to consider the connexion
of the Zopyrus episode with the general course of events, and he omits also
to notice what seems to be by far its more important bearing on the question
of the date when Herodotus composed his work.

It is this point omitted by Kirchhoff that must now be considered.
Herodotus knows half of the story told by Ctesias, but not the whole of it : he
gives us the desertion of Zopyrus, but not his death in the Athenian service.
Now this might well be thought to be a far more significant omission than any
of the others in Herodotus' history on which Kirchhoff lays such great stress.
I cannot think that, if Herodotus had known, when he wrote Book III. 160, the
tragic end of Zopyrus' chequered career, he would have omitted to chronicle
it. It presents an exact parallel to the story of Sophanes at Plataea (ix. 75)
or of the diviner Hegesistratus (ix. 37), in both of which cases Herodotus tells
the story of their deaths, though it has no bearing on the context in which he
introduces them. Other instances could be given, but these are sufficient.

If, however, we suppose that Herodotus left Athens for the West in 440,
it becomes rriuch easier to understand why no record is given of the subse-
quent story of Zopyrus. Moreover a good and sufficient reason can be
suggested why the historian should have started on his travels again just at
this time.

If anything can be stated as certain as to the life and interests of
Herodotus, it is that he had a close connexion with Samos, and a great
affection for that island and its inhabitants. Samos plays a larger part in his
history than any other Greek city except Athens and perhaps Sparta, and the
historian is invariably a ' little blind to their faults,' and ' very kind to their
virtues.' Hence it is surely not carrying conjecture far to suppose that
Herodotus was deeply grieved to see Athens and Samos at deadly enmity,
and his own friend, the poet Sophocles, in command against his former
Ionian home. We may therefore date with some confidence Herodotus'
departure for Thurii as taking place in 440.

It is true that Strabo (p. 656) says that Herodotus ' took part in the
colony to Thurii,' and that Suidas (s.v. 'HpoSoTo?) says he went e? to Sovpcov
airoLKi^ofxevov inrb tow ' AOrjvaccov — ' when it was being colonized by the
Athenians ; ' but even if it were necessary to attach great importance to the
exact words of these authorities — and in the case of Suidas at any rate, the
notice of Herodotus is full of demonstrable inaccuracies — their words are
quite consistent with the view that he joined the colony three or four years
after it had been sent out. No one would hesitate to count John Harvard
among the ' founders of New England,' although he did not sail with the
Pilgrim Fathers in 1620.

The connexion of Herodotus and Zopyrus then may be briefly conjectured
to be as follows. Zopyrus arriving in Athens in 441 or 440 would naturally

THE PERSIAN FRIENDS OF HERODOTUS 45

come into contact with one who like himself had been a Persian subject, and
who knew far more about things oriental than any other Athenian of his
time. We can imagine the historian eagerly drawing from this noble Persian
full details as to official arrangements and as to court secrets, which he had
failed to obtain when himself on his travels in the East. We need only
suppose that they spent some months together at Athens; then Herodotus
sailed for the west, to avoid seeing the end of a struggle between two cities,
both of which he had reason to love, while Zopyrus again turned his face
eastward to meet his death. When Herodotus returned again to Athens,
events had taken quite a new turn; and we can well understand why
Herodotus never completed his story of Zopyrus, even if we accept the
conjecture that he owed to him much important and valuable information.

Before I end this paper, it may be worth while to consider the accuracy
of one important section of the information which Herodotus, as we suppose,
derived from Zopyrus, i.e. the episode of the capture of Babylon which ends
Bk. III. It is unnecessary to recapitulate the well known details in Hero-
dotus as to the desperate resistance of Babylon, the hopeless position of
Darius, and the self-devotion by which the elder Zopyrus saved his king
from a most difficult situation. I propose only to consider the two great
criticisms which are brought against Herodotus' narrative :

(1) It is maintained by many that he has completely misunderstood his
authorities and that he ascribes to Darius a siege which really was carried out
by Xerxes.

(2) The whole story of the self-devotion of Zopyrus is rejected as a fable.
These two criticisms must be discussed separately.

The first criticism is practically that urged long ago by Ctesias : he, we
are told, related of Megabyzus the story told by Herodotus of the elder
Zopyrus. Sayce (ad loc.) seems to attach some weight to the evidence of
Ctesias ; but no one is likely, I think, to be seriously influenced by Ctesias as a
witness against Herodotus or by Sayce as a critic of him.

Other historians who ascribe the siege to Xerxes are Noldeke (doubtfully
in E.B. xviii. p. 572) and Lehmann ( Woch. fur Klass Phil, 1900, p. 963).
The reasons are : —

(1) It is impossible to fit a siege of ' 20 months' (the duration given by
Herodotus iii. 153) into the narrative of the Behistun Inscription.

(2) Lehmann tries to fit in Herodotus' '20 months' with the dates
of Babylonian inscriptions of the time of Xerxes. But his attempt, though
ingenious, will not convince anyone who does not wish beforehand to be
convinced. There are at least two uncertain quantities in his equation.
In fact the evidence from the Babylonian inscriptions is actually used by
Maspero (Hist. Anc. iii. p. 077, n.) on the opposite side to Lehmann,
i.e. to support Herodotus.

(3) The third argument is that the cruelty of the victor (Herodotus iii.
159) after taking Babylon is more in keeping with the character of Xerxes than
with that of Darius.

It will be obvious that of these three arguments only the first is worth

46 J. WELLS

anything. If the Behistun Inscription contradicts Herodotus, no one will
maintain his accuracy against it. But does it contradict Herodotus ? Leh-
mann (ut sup.) and E. Meyer (G. des A. i. 614) say that it does ; Duncker
and Maspero (ut sup.) say that it does not. I will quote the words of the
inscription, (col. II. par. I.) ' says Darius the King. Then Nidintabelus with
the horsemen faithful to him fled to Babylon. Then I went to Babylon. By
the grace of Ormazd I both took Babylon and seized that Nidintabelus.
Then I slew that Nidintabelus at Babylon.'

So far the narrative goes decidedly against Herodotus. Taken by itself
it would seem to imply a speedy capture of the rebel city. But the next
paragraph points as decidedly the other way. ' While I was at Babylon, these
are the countries which revolted against me : Persia, Susiana, Media, Assyria,
Armenia, Parthia, Margiana,Sattagydia, Sacia.' Clearly the siege of Babylon was
a long business. It is not necessary to accept Herodotus' ' twenty months,'
though they may be accurate ; but surely it is unreasonable to reject his whole
story, and suppose that he committed so gross a blunder, and made such a
foolish confusion, as to an important event that happened only some forty
years before his birth.

On the whole then the evidence against Herodotus' accuracy on this
point seems quite insufficient to outweigh the a 'priori probability that he
knew what he was writing about.

With regard to the story of the self-mutilation of Zopyrus, I hope that
I shall not be thought unduly credulous when I say that it seems to me,
though no doubt exaggerated, to contain a s >lid basis of truth.

The arguments against it are : —

(1) It is not mentioned in the Behistun Inscription. If it had been
ever so true, would it have been mentioned ? It was much more creditable
to Darius the king to take towns by the ' grace of Ormazd ' than by the
mutilation of Zopyrus.

(2) But it is urged, no mutilated man could have been set over the
province of Babylon. We need not take Herodotus too literally in his
details; Zopyrus probably made himself ' noseless ' and ' earless' pretty much
in the sense in which

' Earless on high stood unabashed Defoe.'

But I have no doubt he gave himself some permanent scars. Who would
have been offended by these but the Babylonians, whose feelings Darius was
not very likely to spare ?

(3) But it will be urged the story is a well-known legend. Sir H.
Etawlinsorj writes: 'The story told by Polyaenus (and Herodotus) is in its
minutesl features identical with a certain standard oriental tale told by the
bards of Persia, India, and Cashmeer.' But all these stories are long
subsequent to Herodotus, and may well be as much. echoes from his narrative
as is that of Livy as to the self-mutilation of Sextus Tarquinius (i. 54).

Polyaenus tells us that Zopyrus was copying the seltVdevotion of a Sacan
Risaces who had tried to destroy in this way the army of Darius. This story

THE PERSIAN FRIENDS OF HERODOTUS. IT

is quite independent of Herodotus, and may be held to confirm his narrative
as least as much as to refute it.

For the story in its main outlines it may be urged : —

(1) That apart from Herodotus and Polyaenus, it is told by Frontinus
(Stmt. iii. :>, who [nits it in the time of Cyrus) and Justin (i. 10). Ctesias
obviously told the same story, though in his violent antagonism to Herodotus
he misdated it.

(2) That Zopyrus was made ruler of Babylon is an undoubted fact.

(3) If we can accept the story, it suits its context well. Darius was in
a hopeless position, with an impregnable town to capture and an empire
falling into greater revolt every day. The self-devotion of Zopyrus had an
adequate motive and an adequate result.

The second and easy capture of Babylon by Intaphernes (Beh. Inscrip.
iii. 14) is easily explicable. The walls of the town had been breached in all
directions, and it was about as indefensible as Liege in Scott's Quentin
Durward,

I am conscious that in maintaining the accuracy of Herodotus as to the
siege of Babylon, I am distracting attention from the main argument of this
paper. The two points are only partially connected. It is quite possible to
accept the view that Herodotus derived important information from the
younger Zopyrus, even if we also feel ourselves compelled to convict
Herodotus of undue credulity in accepting the whole of his stories.

The first part of my paper I am conscious consists of a series of
hypotheses. In the fragmentary state of our evidence, no other method of
inquiry is possible. I hope, however, that some of them may be thought to
throw light on a difficult and important subject,

J. Wells.

THE FLEETS OF THE FIRST PUNIC WAR.

According to Polybius, there took part in the battle of Ecnomus 680
quinqueremes and 290,000 men, i.e. crews 204,000 and troops 86,000 ; while in
the next year, at the battle of the Hermaean promontory, 550 quinqueremes
were engaged. The only figures comparable to these in Roman history,
manifest absurdities apart, are those given by Appian for the battle of
Naulochus, and perhaps those for Actium. At Naulochus 300 ships of all
sizes are said to have been in action on either side, and no doubt Agrippa's
fleet, at any rate, did amount to this large number 1 ; while at Actium
Octavian may have had anything up to 400. 2 But in Octavian's time the
population of all Italy may have been 7 to 8 millions 3 ; the Mediterranean
was almost a Roman lake, and its entire resources went to furnish the fleets
for the civil wars. But for the generation next after that of the first Punic
war, the population of Roman Italy has been reckoned at only 4 to 4|
millions, that of Carthaginian Africa at perhaps 3 to 4 millions, 3 while the
Mediterranean supported several considerable fleets beside those of Carthage
and Rome. More than one writer has seen that the numbers given for
Ecnomus are impossible 4 ; and it seems worth while making an attempt to
get at some more reasonable figures for the first Punic war.

As to the materials, if Polybius is to be corrected it must be from Polybius
himself, and not from the later writers. 5 Apart from his being a great
historian, he is far nearer in time to the original tradition than any one else.

1 J. Kromayer (' Die Entwickelung der rom- the figures for the Actium campaign are very
ischen Flotte,' Philologies 1897), who has gone uncertain.

into the figures for the civil wars, accepts 300 for 3 See J. Beloch, Die Bcvolkerung der gr.-

the fleet of Sextus Pompey also. But this seems rom. Welt; also Die Bevblkerung Italiens in

to me impossible ; for Pompoy's 3 squadrons at Allerthumm Beitrage zur alien Geschichte,vo\. 3.

Mylae, totalling 155, are described by Appian 4 The following helps one to realise what

instituting the larger part of his fleet ; such figures mean. On a population of 42

after losing 30 at Mylae and some at Tauro- millions, the British Navy has a personnel of

menium he cannot have had more than 250 at 121,983 (including coastguard and marines),

the most al Naulochus, for building between the and mobilised 319 vessels of all sorts for the

two battles was out of the question. This manoeuvres of June — July, 1906 ; while in crew

would give a total of about 550 ships in and troops two quinqueremes carried about the

action. same number as one battleship.

2 J. Kromayer in Hermes 34(1899) p. 1. If 5 Meltzer has stated this(Gesch. der Karthagcr,
Octavian had 100, and Antony 170 (plus 60 vol. 2, p. 568, n. 49). But he makes no
Egyptian), over 600 ships were engaged. But application of it to the numbers.

THE FLEETS OF THE FIRST PUNIC WAR. 49

Of the rest, Zonaras (Dion Oassius) is confused and gives no figures. The
epitomators of Livy, as they often disagree, must be the subject of consider-
able textual corruption; but even could we restore Livy, he must either
agree with Polybius or be of less authority. There remains Diodorus. It
seems agreed that, while Polybius is partly Fabius, partly Philinus, and partly
neither, Diodorus is certainly largely Philinus, i.e. that lie often gives what is
substantially the Carthaginian version. Now I regard it as certain that
Philinus would tend to exaggerate the Roman numbers, for obvious reasons,
just as Fabius would the Carthaginian; Diodorus may therefore be of
occasional use as giving a superior limit for Roman figures. I assume
that, other things being equal, the smaller of two numbers is to be taken.

What, now, was the position when war broke out ?

Carthage had finally got the better of Syracuse in their secular duel, and
was the greatest sea-power of the wesi. But it is easy to exaggerate to
oneself that power. Meltzer gives an instructive list of prior Carthaginian
fleet-numbers: 4<S0 B.C., 200 warships; 406 B.C., 120 triremes; 397 B.C. (war
with Dionysius I.), 100 triremes, raised to 200 the next year ; 368 B.C.
(again against Dionysius), 200 warships; in Timoleon's war, first 150, then
200 warships ; in 311 B.C., 130 warships ; finally, 130 offered to Rome for help
against Pyrrhus. (I omit two small squadrons prior to the fourth century.)
These numbers are chiefly from Diodorus, and may not be accurate ; but
anyhow they shew two things ; first, that there was a tradition that in a time
of supreme national effort Carthage could raise a fleet of 200 ships ; secondly,
that it was believed that the ordinary establishment, of the Carthaginian fleet
prior to the war with Rome was 130 or thereabouts. Whether these two
beliefs existed at the time of, or whether they were a consequence of, the first
Punic war may for the moment be left undecided.

Rome, of course, had possessed, or had had the control of, warships since
the middle of the fourth century B.C., if not earlier. 6 But a distinction must
be made, prior to the war with Carthage, between the true Roman fleet {i.e.
the duumviral squadrons) 7 and the ships which, after the war with Pyrrhus,
were liable to be furnished under treaty by the Italiot towns. Duoviri
navales were first created in 311 B.C.; in 283 B.C. a squadron of 10 ships
under a duovir was attacked by the Tarentines and five ships taken s ; in 181
B.C. 9 and. 178 B.C. 10 we find that the double duumviral squadron consisted, on
each occasion, of 20 ships, each duovir commanding 10. We may perhaps
assume that the double duumviral squadron was regularly 20 ships.
Such a squadron was only fitted out when required, and then laid up again.
Polybius says that when Appius Claudius crossed to Messana the Romans

6 The view that Rome, prior to 260 B.C., used with Carthage, which cannot he later than

Greek ships only, seems again coming into 348 B.C., presupposes Roman warships,

prominence, no donht as a reaction again si " Not, of course, manned by Romans.

Mommsen; see aj. K. Speck, Handclsgcschichle, 8 Liv. Per. 12 ; A]']'. Samn. 7, 1.

3, i, §715. lint it is demonstrably wrong. * Liv. l<>, 28.

Rome controlled no Creek ships before :'.'J7 B.c. "' Liv, 11, 1.
(treaty with Neapolis) ; while the first treaty

U.S. — VOL. XXVII. I

50 W. W. TARN

had not a single ship of their own ; no squadron therefore had been fitted
out. But probably in the navalia were at least 15 old ships, the remains of
the squadron of 283 B.C. 11

They had, however, afloat a squadron of triremes and pentekontors,
furnished under treaty by various Italiot towns, 1 ' 2 Tarentum, Locri, Elea,
Neapolis. These treaty contributions were very small. In 210 B.C. D.
Quinctius obtained 12 ships from ' the allies and Rhegium and Velia and
Paestum ' in full discharge of their obligations (Liv. 26, 39). In 191 B.C.
C. Livius got from Naples, Rhegium, Locri and the socii ejusdem juris 25
open ships due under treaty, some being rostratae, some speculatoriae (scouts). 13
The treaty contribution of Carthage herself in 191 B.C. was only six
cataphracts (Liv. 36, 4 and 42). Messana, says Cicero, had to supply one
ship. The obligation of Rhegium was one ship, that of Locri 2 (Liv. 42, 48).
We cannot suppose that the Romans got more than 25 ships from the Italiot
towns in 260 B.C.

The Romans, having resolved to contest the sea, built 100 quinqueremes
and 20 triremes. The 20 triremes must represent the double duumviral
command, 14 the number the Romans had previously been accustomed to
build when they wanted a fleet. The 100 quinqueremes are probably correct,
seeing that the first measure of the Romans, when war broke out with
Antiochus, was to decree 100 quinqueremes, 15 though they were never all
built. The Romans must also have refitted any old ships in the navalia,
their regular operation at the beginning of a war {e.g. Liv. 35, 20; 42, 27).
The Roman fleet therefore would consist of 120 ships newly built, some 15
refitted, and some 25 Italiot ; possibly also two or three from Massilia 16 ;
that is to say, about 160 altogether. 17 Obviously, Rome was not going to
challenge Carthage with deliberately inferior numbers, though an exaggerated
idea of the strength of the Carthaginian navy has led most writers to suppose
that she did so ; the Carthaginian fleet in 260 B.C. should therefore be
somewhat less, and no doubt the number was 130, the number which
Polybius gives them at Mylae (possibly taken from Duilius' column), and
which agrees with, or else was the cause of, the already noticed belief that

11 Mommsen thinks the Tarentines had to give 15 Liv. 35, 20. In the affairs of Greece and
up their ships after the war with Fyrrhus ; but Syria Livy is supposed to represent the
probalily this was not the case (see Niese, substance of Polybius fairly accurately, and for
Grundriss d. ram. Gcsch. in Midler's H. d. k. the naval war with Antiochus the way in which
A. Ill, •'), (1906), ]>. 70, n. 4), for Tarentum Appian agrees with and complements Livy
kept its independence and had a number of ships makes this almost certain. If the 100 quin-
Ln the second Punic war. Even if they did hand queremes of 260 B.C. be from Fabius those of
over some ships, the Romans, as was their 192 B.C. are not.

custom, probably burnl them. 16 Two Massiliot ships joined Cn. Scipio in 217

12 Polyb. 1, 20. H.c. (Polyb. 3, 95 = Liv. 22, 19) ; and in 211

13 The number appears (quite clearly) from p.. c. four joined the propraetor M. Junius Silamis
App. Syr. 22 combined with Liv. 36,42, and is (Liv. 26, 19).

presumably that of Polybius. 17 Naturally I attach no importance to the

14 I do not mean that they had anything to fact that Floras 1, 18, 7 says the Romans built
do with^duoviri, who are not heard of during 160 ships in 260 B.C.

bhe period "i tin- great naval war.-..

THE FLEETS OF THE FIRST PUNIC WAR. 51

the ordinary establishment of the Carthaginian fleet prior to the war with
Rome was about 130.

As regards the opening of the naval campaign, it is clear that in chs. 21
and 22 of book 1, Polybius has combined two different accounts. 18 Both com-
menced with Boodes capturing 17 Roman ships; ch. 21 then makes Hannibal
blander into the Roman fleet with 50 ships and lose some 30 (' more than
half); butch. 22 knows nothing of this; here the main Roman fleet, still
far off and concerned at Boodes' victory, puts in (? to Rtessana) and equips
itself with the corvus. The account of Mylae that follows, the Carthaginian
confidence, the honours paid to Duilius, are all inconsistent with a prior
Carthaginian defeat; and the battle of ch. 21 must undoubtedly be, as Dr.
Beloch supposes, the Carthaginian version of Mylae taken from Philirrus, 19
though Polybius may well be excused for not recognising it. If so, it is some
evidence that (as we may, indeed, suspect from Polybius) the whole
Carthaginian fleet was not engaged in that battle.

The Roman fleet at Mylae, then, was some 140 strong, (about 160 less 17,)
against the whole or part of a Carthaginian fleet of 130. The Romans
took 30 ships and the hepteres, and sank 19. Their own loss is not given,
but must have been less than 19 ; say 10. If they were able to refit 20 out
of the 30 prizes," 20 they were probably about 150 strong the next year.

Hannibal, with the 80 ships left, returned to Carthage, procured rein-
forcements, (probably Boodes' prizes and a few old ships,) and sailed to
Sardinia perhaps 100 strong. Here he lost ' many ' ships, but apparently not
their crews. As Polybius gives the total Carthaginian loss in the war at
about 500, we can see, by adding up the other losses in his figures,- 1 that he
must have taken the loss in Sardinia at about 60. But it will appear that we

18 F. Reuss, Philologus 60 (1901 j p. 102, who grapnel on a pole, like Agrippa's ap?ra|. Dion

has made the latest examination of Polybius' Cassius so understood it, for Zonaras speaks of

sources, gives Chs. 20-24 as all from Fabius. the Romans using x«'P a * -ntpiKovTovs aiSrjpas :

But such a result seems to me merely to and Appian so understood it, for whereas

condemn his method ; for the break in the Agrippa's ships at Xaulochos use the apira^

sense of the narrative between Chs. 21 and 22 (described App. 6. c. 5, 118), at Mylae these

is paten 1 to anyone. same ships have xopaKas (b. c. 5, 106). The

1:1 Gr. Geseh. vol. 3. i. p. 677 n. 1. If this supposed boarding-bridge must have been

be so, it is noteworthy that Philinus knows taken by Fabius from the bridge or ladder of

nothing of the boarding-bridge (corvus, K<fpa|) ; the sambuca used by Marcellus in attacking

ami no doubt Ihne was right in suspecting the Syiacuse, eo doubt a real machine, as it appeal's

traditional account of this machine, which is again (somewhat altered) in the siegi> of

not heard of after Ecnomus, and which seems Rhodes and Cyzicus by Mithradates (App.

part of the deliberate introduction by Fabius Mith. 26 compared with 73 . Some sort of

<>t in element of wonder into this war: for. ladder for boarding occurs, however, at the

after all, boarding and irffanaxia were the battle of Cumae (App. i.e. 5, 82), if this 1

oldest form of sea-fighting known, and the Car- meaning of Appian's unique use of KarapfraKTai.

thaginians would have been delighted with an '-'" About the proportion refitted after

arrangement that would have prevented more Ecnomus.

than two Romans coming aboard at once. Now '- 1 Viz, battle of ch. 21, 30, Mylae 50,

the Athenians had used grapnels in 413 B.C., Tyndaris 18, Ecnomus about 100 (64 + over

and they occur commonly in the second Punic 30), rlermaea 111, Aegatcs Insula.- 120; 432

war; and the Ktfpaf was probably an improved altogether.

E 2

52 W. W. TARN

require some further loss for Hermaea ; and 500 is a very round figure. We
may put Hannibal's outside loss in Sardinia at 40; it may have been nearer
20, leaving him some 60 to 80 ships.

These ships encountered the Roman fleet at Tyndaris (257 B.C.). That
the Romans were in greatly superior numbers (we have seen it might be
about 150) appears from the account of the battle 22 ; and the only extant
figures 23 are at least evidence of a great disproportion in strength. The
Romans lost nine ships, the Carthaginians 18.

The Carthaginians had paid the penalty of despising their enemies.
They now set to work in earnest to beat them, as did the Romans to invade
Africa; both, says Polybius, made a great effort. The results were Ecnomus
(256 B.C.) and Hermaea (255 or 254 B.C.). 24 The figures in Polybius are as
follows : Ecnomus, Romans 330, Carthaginians 350 ; Roman loss 24 sunk,
Carthaginian more than 30 sunk, 64 captured. Hermaea, Romans 350 (i.e. 330
less 24 sunk plus 44 prizes refitted, 25 the 40 ships left in Africa taking part
in the battle), Carthaginians 200, some of which had been built in a hurry
(Polyb. 1, 36); the Romans capture 114; no other losses given. On the
way home the Romans encounter a storm, and out of the 364 all are lost
but 80.

Here are two big discrepancies. If the Romans had 350 ships at Hermaea,
then (on Polybius' evident assumption that they had no losses) they should
have had 464 ships in the storm, not 364, (i.e. 3504-114 prizes in tow) ; while
the Carthaginians, with 250 left after Ecnomus, need not have built in a
hurry to get 200 to sea.

To take the Roman figures first. Supposing Polybius' account of
Hermaea to be correct, the figure 364 for the storm (2504-114) shews, on
the assumption of no Roman losses, that the Roman fleet at Hermaea was
250. 20 If so, that at Ecnomus was 230 (230-24 sunk + 44 prizes refitted,
as before = 250). Is, then, Polybius' account of Hermaea correct ?

~- 2 The consul hurries after the enemy as an different proconsular squadron-commands of

easy prey, going forward with 10 ships ; they the 2nd Punic war can hardly be cited in

surround him and sink 9, but, pursuing the support.

flagship, become engaged with the main Romrui ,& Polyb. 1, 29, says they refitted the prizes,

fleet, and lose 8 sunk, 10 taken. If he means all, which is unlikely, then only

21 Polyaen. 8, 20 ; Romans 200, Garth. 80. forty-four were taken, and the lower of the two

i4 The date is fortunately not material here, numbers hereafter discussed for the Cartha-

for either year is open to serious objection. ginian fleet at Ecnomus becomes even mine

For a summary of the argument", see Reuss, probable.

u. s. ; also Beloch, Gr. Gesch. 3, ii, 234, whose - 6 The number 350 given for Hermaea does not

reasons for 254 are hardly convincing. The shew that the storm number should be 464,

difficulty is this : 255 f;ives no time for the because, ceteris paribus, the smaller number is

siege of Clypea, and does not explain why the to be followed. But as a fact Eutronius, or

Hermaea triumph Fell in 253; while 254 makes some scribe, had the curiosity to add up, and

the Romans first waste a year before succouring Eutropius does give 464 for the storm; and

their beaten troops (though the fleet was ready), Meltzer hereupon suggests that the real Foly-

and then send out the fleet, not under the bian tradition may have been 464, a suggestion

consuls (as on all other occasions in this war), which is out of the question for at least three

but under the consuls of 255-4 as proconsuls, separate reasons : it prefers the easier version

without any apparenf reason ; for the very and the large!" number, and corrects a good

THE FLEETS OF THE FIRST PUNIC WAR. 53

It has been criticised on two grounds : one, because Polybius gives 114
Carthaginian ships as captured ami none as sunk ; the other, because he
dismisses in three lines what (on his shewing) was a greater victory than
Ecnomus, to which he gives as many chapters. There was, too, another
(? Carthaginian) version of this battle, which makes it a stubborn fight
(Zonaras), the Carthaginians losing 24 slups (Diodorus); and I line was
inclined, following Haltaus, to take Poly bi us' eKarhv SeKareaaapa^; as a
corruption of ei/coai /cal reaaapa^.

Correcting Polybius' account of a Carthaginian loss by Diodorus is
hardly convincing work; but in fact there is little doubt that the battle was
a great defeat for Carthage. Not only was she impotent at sea for years
after, but the consul Aemilius Paullus, who was in command, set up a
columna rostrata to celebrate the victory (Liv. 42, 20), and we only hear of
one other such column prior to Augustus, that of Duilius. As to the
captures, Polybius' phrase e£ icpoSov /cal pahLws rpe\frd/xevoi shews that the
battle was of the Drepana type : the Carthaginian fleet, in part hastily built
and manned by crews of whom some must have been inexperienced and the
remainder possibly shaken by a great defeat, was surprised or caught at a
disadvantage and jammed against the shore, all, or almost all, the ships that
could not make the open sea being captured. 27 And Polybius presumably
dismisses the battle in three lines just because he had given so much space
to Ecnomus, for he had to keep his account of the war brief. 28

The Roman numbers, then, are 230 Ecnomus, 250 Hermaea, 250 + 114 in
the storm, of which all were lost but 80. 29 The number 330 for the Roman
fleet at Ecnomus no doubt arose from reckoning in the transports 30 and
calling the whole warships; the number 350 for the Carthaginian fleet merely
shews that Fabius, as a good patriot, had given a number a little bigger than
that of his own side as he made it out, The hurried building of the
Carthaginians before Hermaea may have been from 50 to 100 ships, according
as from 150 to 100 escaped from Ecnomus; the figure, then, at Ecnomus
would have been at the outside 250 (100 being lost), but might not have
exceeded 200. Apart from the preference to be given to the smaller number,
if sufficient, other considerations all point to 200. The Roman number 230
shews that they expected to meet a fleet of not over 200, or else, looking to

early writer with a sound text by a poor and which Livy gives at length ; and this in a war

late compiler. Reuss, u. s. and Speck, Han- where his general agreement with Livy over

delsycsch. 3, ii, § 824-5, follow Meltzer ; but the naval operations is most marked,

one cannot write history merely by taking the ~ v If any Roman ships were lost at Hermaea,

line of least resistance. the number lost in the storm would be fewer,

- 7 Possibly the number 114 comes from the SO anyhow remaining.

column. Perhaps, too, the reason why Philinus 30 This must be a common source of confusion

(Diodorus) gives the Roman loss at Drepana in classical as in modern) fleet numbers. No

as 117 was to shew that the Carthaginians figures in antiquity are more exaggerated than

had had a full revenge for Hermaea. those of transports; after such numbers as

28 There is an exact parallel in Appian's 3,000 and 1,600, the writer who confined him-

account of the war against Autiochus; he dis- self to less than four figures must have been

misses in two casual lines the very important astounded at his own moderation.
defeat of Hannibal by the Rhodians at Side,

54 W. W. TARN

what they did later, they could easily have built more, having some 140 ships
and 10 prizes in hand to start with. The Roman number 250 at Hermaea,
which came automatically without building - , would have been increased had
Carthage shewn ability to put 250 to sea, and almost proves that the
Carthaginian number at Ecnomus was less ; for it is to be remembered that,
both before and after Ecnomus, Rome, in addition to her greater resources,
had a very long start in building. We have, too, the tradition, whether prior to
or due to this war, that 200 ships meant a supreme effort for Carthage. 31
Most important of all, perhaps, is the battle itself, which points to the
Carthaginians being outnumbered; they tried enveloping 82 tactics, and failed
because their centre was too weak for its work. We must, I think, give
Carthage at Ecnomus 200, as at Hermaea ; anyhow not much over. If the
Romans after Hermaea took off 114 prizes in tow, there must have been a
few ships too badly wounded to tow 33 ; if we say 16, and give Carthage some
70 not very efficient ships remaining, that is all they can well have had. 34

To continue Polybius' figures. After the first storm, off Camarina, the
Romans, having 80 ships left, built 220, raising their fleet to 300 (254 B.C.) ;
they capture Panormus (253 B.C.) ; they lose 150 ships in a second storm, off
the Lucanian coast (253 B.C.), and retire from the sea; in 252 B.C. they escort
a convoy to Panormus with 60 ships; they again build 50 ships, making
200 in 250 B.C., in which year they form the siege of Lilybaeum ; in 249 B.C.
P. Claudius has 123 ships at Drepana, and L. Junius 120. In 251 B.C.
Hasdrubal sails to Sicily with 200 ships and a large army ; after Drepana
Adherbal receives a reinforcement of 70 ships under Carthalo. These are
all the numbers given by Polybius. It will be best to work backwards from
Drepana.

Claudius' plan was to sail from Lilybaeum to Drepana with every ship
he had. 35 Polybius says 30 escaped, and the rest, 93, were captured ; the
account shows that some of them were much damaged. His fleet, then,
numbered 123. Adherbal's force is not given. It must have been smaller
than the Roman ; first, because Claudius thought it feasible to attack him
under the catapults of Drepana; secondly, because Adherbal's victory was looked
upon as an unexpected salvation for Carthage ; thirdly, because in Polybius'
list of the advantages on the Carthaginian side that of numbers is not
included. At the same time, it was large enough to capture the bulk of
the Roman fleet. We shall not be far wrong if we put it at 100 at the
outside, possibly rather less. Why Claudius attacked is clear enough ; he
must have heard that Adherbal was about to receive a reinforcement of
70 ships, (which in fact arrived after the battle,) and he very properly

31 The difficulty, of course, all through (money 34 Assuming that they did have as many as
apart), both at Rome and Carthage, must have 200 ships at Hermaea.

been, not ships, but men to row them. 35 Polyb. 1, 49, navri r$ aT6\<f. I mention

32 The Romans are described as in wedge this as it is commonly assumed that he left a
formation, not in line. squadron at Lilybaeum, a most useless proeeed-

" Even at Drepana, Polybius says, there was ing, as the blockade was a failure anyhow, and

some ramming, ami sonic ships settled. bis striking force insufficient.

THE FLEETS OF THE FIRST IT NIC WAR. 55

supposed that if he did not attack while he could, that able man would
presently attack him in overwhelming force. No wonder he lost his temper
with the sacred chickens.

After the battle, Carthalo, with the 70 ships he had brought and
30 others 36 given him by Adherbal, attacked the 30 Roman ships that had
escaped to Lilybaeum with Claudius, and accounted for ' a few ' (dXiya) of
them, towing off some and destroying others. Diodorus says he captured 5 and
sank a few; and Philinus would make the most of it. Putting both accounts
together, we may say that Carthalo cannot have accounted for more than 10
of the 30. Carthalo then took up his station not far from Lilybaeum, to
hinder the approach of the other consul, L. Junius," 7 who was coming up with
a convoy and 120 warships, which figure included ships that had joined him
from ' the camp and the rest of Sicily.' 3S He had these 120 before < Jarthalo's
attack on the thirty ships at Lilybaeum, and anyhow the surviving 2(J could
not have joined him, as Carthalo with 100 ships lay between. Junius'
entire fleet was lost in a storm. At the end, then, of this disastrous year,
in which the Romans lost some 223 ships, they had some 20 only remaining.

Now to work backwards. The Romans built 50 ships in 250 B.C. ; in
249 bc. they had 243; their number, then, in 250 B.C., before they built, was
not 150, as Polybius says, but 193. They did not, therefore, lose 150 ships in
the second storm, off the Lucanian coast. Now they had 80 ships left after
the first storm, and are said to have built 220, making 300 altogether. Why
they should raise their fleet to this unparalleled figure at a time when
Carthage was quite impotent at sea does not appear. Diodorus gives the
total Roman fleet after this building (not the new-built ships only) as 250,
and we have assumed that Philinus was likely to exaggerate the Roman
strength. The real number, therefore, was probably under 250; and as we
have to account for the figure 220 in Polybius, there can be little doubt that
220 was the total, not of the newly built ships, but of the whole Roman
fleet after the building. 39 The fleet, then, in 254 B.C. was 220 ships ; the loss
in the second storm in 253 B.C. was not 150, but 27 (220 — 27 = 193) ; 193,
with the 50 built in 250 B.C., make up the 243 required for the year of
Drepana. The supposed loss of 150 in the second storm must, then, be a
duplicate of the loss in the first storm ; 40 and if the Romans sent only 60

" ; Polybius says the prizes were taken to him from Lilybaeum, they must have been sent

Carthage. Some were damaged ; if we take the off before the battle of Drepana.

same proportion as after Ecnomus, CO t> 70 39 The 300 of Polybius may be another

at the most would be worth refitting, and instance of confusion due to transports or ships

Adherbal would have remaining jusl about the other than warships ; for Diodorus says the

same number of ships to tow them. Romans sailed to the siege of Lilybaeum with

37 The Fasti shew that he was Claudius 3 240 long ships and 60 cercuri, i.e. 300 vessels,

colleague. Polybius speaks as if he were his Cercuri occur in the Roman navy (or the Roman

successor and sailed the next year (248 b.c. . service) in both the second and third Punic wars ;

but it seems reasonably clear that the naval I.iv. 23, 34 : App. Lib. 75.

operations under both consuls form one 4 " At first sight the loss in the first storm

connected sequence and took place in the same would seem to be 170 Roman ships and 114

year, 249 B.C. prizes; for the prizes would, of course, have

33 Pol. 1, 52. If, in fact, any ships joined been casl off when the storm broke. But as

56 W. W. TARN

with a convoy in 252 B.C. it was because GO sufficed, and not because they
were retiring from the sea.

For that the Carthaginian navy did not easily recover from the battle
of Hermaea seems certain. They had perhaps 70 not very efficient ships left,
and the Romans did what they liked at sea. They took Panormus ; they
sent supplies there with only 60 ships as escort; they blockaded Lilybaeum.
The Carthaginians in 250 B.C. could not attempt to raise the blockade ; they
had to confine themselves to running it ; how little there was to fear from
the sea side is shown by the Romans dividing their fleet, and also laying up
part of the blockading fleet and using the rowers as land troops. The
history of the second Punic war seems to shew that Carthage could not, and
knew she could not, support at once a great fleet and a great army ; and in
251 B.C. she had sent to Sicily the army and the elephants with which Has-
drubal attempted to retake Panormus by land. 41 The destruction of that
army and the danger to Lilybaeum, however, compelled Carthage once more
to turn to her fleet; by 249 B.C. Adherbal had perhaps 100 ships, and 70
others were ready at Carthage ; it was this growing danger that compelled
Claudius to strike.

For the period after Drepana there is little to say. The Cartha-
ginians had some 170 ships, less any lost at Drepana, plus some 60 to 70
prizes worth refitting. But after a little they laid up their fleet, no doubt
because the crews were wanted for the war already on their hands in Africa ; i2
under these circumstances it is wholly unlikely that any prizes were fitted
out. When the Romans again built they built 200 ships ; these, with the
20 or so remaining after Drepana, which according to Zonarashad meanwhile
been used as privateers, would give them about 220 in the final battle of the
Aegates Insulae. The Carthaginian number is unknown ; Polybius merely
says they got ready ' the ships'; if we assume that they had 100 ships at
Drepana and no losses there, and could and did refit 70 prizes — all the mo^t
favourable hypotheses, in fact — they may have controlled 240 ships, as an
outside number. But if they laid up their own fleet, it is unlikely that they
had fitted out the prizes. They had used up their trained crews ; both the
rowers and the marines who took part in the battle were extemporised ; no
doubt they were in part got together from the crews of the transports ; for
that there were no men to spare for transports is shown by the warships them-
selves being loaded down with stores for the army of Sicily. It is not in such

rostra were forthcoming for Aemilius Paullus' to this that Hasdrubal did get his large army

column, one division must have got to harbour across ; the Romans had no naval base facing

with its prizes ; consequently the Roman ships Africa, and even if they had had, the command

lost were more than 170. of the sea (such as it was with galleys) rarely,

u The 200 'ships' with which Hasdrubal if ever, prevented an army crossing in ancient

dtoSicily Polyb. 1,38) are obviously trans- times. Pompey commanded the sea absolutely

ports. To suppose that they were warships as against Caesar; so did the liberators as

makes nonsense of tin- events before Lilybaeum against Antony and Octavian ; yet in each case

in 250 and 249 B.C., more particularly of the the Adriatic was crossed in force.
Romans laying up pari of their fleet, the account 4 - Meltzer, vol. 2, p. 336.

of which is yery circumstantial. It is noobjection

THE FLEETS OF THE FIRST PUNIC WAR. 57

circumstances, and with such a dearth of men, that they could have got to
sea a fleet of 24*0 ships, the largest in their history. If we give them their
own ships, 170, we shall be nearer the mark, with perhaps 200, the number
of Ecnomus and Hermaea, as an outside figure. 43 But this time seamanship
was on the side of Rome ; 120 Carthaginian ships were sunk or taken ; and
the war was over.

If any reader has had the patience to follow the foregoing analysis, he
will already have seen the deduction from it; but for clearness' sake I may
repeat the figures that seem probable. In 200 B.C. the Romans had about
100 ships, the Carthaginians 130. At Ecnomus, Romans 230, Carthaginians
about 200 (with a possibility of a somewhat higher Carthaginian figure). At
Hermaea, Romans 250, Carthaginians 200. In 254 B.C., Romans 220,
Carthaginians about 70. In 249 B.C., Romans 243, Carthaginians about 170.
At the Aegates Insulae, 241 B.C., Romans about 220, Carthaginians perhaps
170 to 200 at the outside.

The tradition, then, that a fleet of 200 ships meant a supreme effort for
Carthaoe dates from before the war, and was well founded ; it was known to
the Romans ; and the Romans, in their bid for sea-power, were not invading
the realm of miracle, but were acting on a reasonable, cool-headed calculation.
They reckoned that, with their greater resources, they could keep up a fleet
of from 20 to 40 ships in excess of 200, that is, in excess of anything they
expected Carthage to do; and that if they did this they must win. And
they did win ; though their calculations were nearly upset by the genius of
Adherbal and the jealousy of the sea. Their victory was none the less a
heroic achievement because it was founded in a well-reasoned policy and
because the Carthaginian sea-power was perhaps not so great as we have
been accustomed to think. 44

One other conclusion appears to follow from the figures. The Romans
were throughout building to the Carthaginian numbers, not vice versd. This
does not necessarily mean that they could build more quickly, for they had
(so to speak) the whip hand in the matter of building from Mylae to
Drepana ; but it does mean that they must have known a great deal more of
what was going on at Carthage than the Carthaginians knew of what was

43 There is another, perhaps a better, way of cent., or even 60 per cent., and we come back

getting at the Carthaginian fleet of 241 B.C. to this, that a fleet of not over 170 cannot be

In the war with the mercenaries the Carthagin- far from the mark. Of course, if the 120 ships

ians had nothing but triremes and pentekontors lost were not all quinqueremes, the argument

(Polyb. 1, 73) ; they had therefore lost all their is even stronger.

quinqueremes at the Aegates Insulae. including " If 200 ships or so was in fact Carthage's

presumably the Roman prizes. Suppose all the effective limit, the limitation must have had to

120 ships lost to have been quinqueremes, the do with the crews, of which we know little. It

swifter ships alone escaping. 120 is 50 per has nothing to do, for instance, with the numbei

cent, of 2-10, (50 per cent, of 200, 70 per cent. of the pewpta at Carthage being 220 : lor. apart

of 170, 80 per cent, of 150. Bui we know that from I'tica. the Carthaginians had the control

in 219 B.C. the Carthaginian fleet of Spain con- of the docks built by Agathocles at Hippagreta

taiued SS per cent, of quinqueremes, (post, (App. Lib. 110) ; and besides, a fleet could at a

p. 9); it is therefore most unlikely that their pinch winter ashore anywhere {e.g. Liv. 30.

fleet of 241 B.C. contained as few as 50 per 45).

58 W. W. TARN

going on at Rome. Did Hieron provide for the intelligence of his allies, as
well as for their commissariat ? 45

The probability of the correctness of the view which I have taken is
much enhanced by a consideration of the figures handed down for the
second Panic war. I am not going into these in detail, but I may give a few
salient points. The Romans began operations in 218 B.C. by sending out
220 ships. 46 By 217 B.C. it was clear that Carthage was not going to fight
at sea. In 215 B.C. the Carthaginians had 120 ships at sea, 47 plus a few in
Spain, possibly 18. 4S In 214 B.C. the Roman fleet is down to 185. 49 In
212 B.C. the largest Carthaginian fleet of the war, 130 ships under Bomilcar,
attempts to relieve Syracuse. 50 In 211 B.C. the Roman fleet is raised to 215,
a new squadron of 30 being fitted out and sent to Spain under M. Junius
Silanus, 51 giving 100 for Sicily, 65 for Spain, and 50 for the Adriatic; the
latter squadron, however, was tied to watching Philip, and could hardly be
counted as available against Carthage. In 208 B.C. there was a scare of a
great Carthaginian fleet, the number, of course, being put at 200. 52 Rome had
already 233 ships this year, i.e. those of 211 B.C. plus 18 taken by Scipio at
New Carthage and fitted out 53 ; all the ships in Spain, however, had been
laid up and the crews added to the army, while the fleet of the Adriatic did
not count as against Carthage. Scipio was therefore ordered to equip and
send to Sardinia 50 ships, and 50 additional ships were fitted out at Rome,
giving, with the fleet of Sicily, 200 ships 52 ; while Silanus had in addition 30
quinqueremes in Spain for which he had crews, and which were available
should the Carthaginian fleet materialise. 54

The events of the year 208 B.C., in which Rome equipped 230 ships to
meet a threatened Carthaginian fleet of 200, do appear entirely to support
the conclusion come to with regard to the first Punic war.

One word as to the total losses given by Polybius. Assuming that his
figures for the losses in the separate battles are correct — and without this
assumption we cannot go into the figures at all — the total of 500 given for
the Carthaginian loss is not very wide of the mark ; as worked out in this

46 The only time, except in Spain, when a out from Livy's confused narrative (24, 36 ; 25
Roman squadron seems to have been outnum- 25 ; 25, 27) whether there were or were not
bered by a Carthaginian was after Syracuse had 20 other Carthaginian ships in the harbour of
joined Carthage ; Marcellus, in 212 B.C., had to Syracuse with Epicydes.

offer battle to Bomilcar with an inferior force. 51 Liv. 26, 19.

4,1 Polyb. 3, 41=App. lb. 14 = Liv. 21, 17. 52 Liv. 27, 22. The SO ships of Scipio that

Livy adds 20 celoces. he mentions are the original 35, the 18 taken

47 Liv. S-), 32. at New Carthage, and the 25 taken from

48 Tin- number captured at New Carthage, Himilco and never fitted out.
Polyb. 10, 17. ss p i yb . 10 , 17.

49 i.e. 150 for Sicily and the Adriatic (Liv. 5I In estimating this tremendous effort (280
24, 11). plus 35 in Spain, the original squadron ships), it must be remembered that the Romans
of Cn. Scipio. There were also in Spain 25 were now in part using slave rowers (Liv.
prizes taken from Himilco (Polyb. 3, 96 = Liv. 24, 11 ; 26, 35); and that some of Scipio's
22, 10}, but it appears from Polyb. 10, 17, that crews were pressed Spaniards (Polyb. 10, 17 =
these were not fitted out. Liv. 26, 47.)

Liv. 25, 27. I find it impossible to make

THE FLEETS OF THE FIRST PUNIC AVAR. 59

paper, the actual loss may have been something like 450. 55 But the total of
Tim) for the Roman loss is much too high, even on Polybius' own showing ;
for it includes the 114 prizes taken at Hermaea, which are thus reckoned
twice over. The Roman loss cannot well have been much over 500. 56 Even
so, these are very large figures ; as large as for the eighteen years preceding
and including Actimn.

Even, however, if the numbers arrived at in this paper seem more
probable than the traditional ones, there still remain two difficulties — the
question of light craft, and Polybius' use of the word Trevrrjpris. The Roman
figures are, of course, inclusive totals, comprising all ships under Roman
control, Italiot or otherwise : it seems that the Romans did not call on
Hieron's navy at all. But a question arises whether the fleet numbers do or
do not include light craft; also whether in the third century B.C. light craft
took part in fleet actions at all, as they undoubtedly often did in the second.
I am not going into this here ; but light craft (by which I mean lembi and
other ships smaller than pentekontors) raise many difficulties in studying
ancient fleet numbers, and may be responsible for many apparently purpose-
less exaggerations. 57 The fleets of the first Punic war were of course
accompanied by a few scouts, 58 but whether these be included or not, they
would be much the same for both sides, and would not alter the proportions.

It is necessary, however, to refer to Polybius' use of irevr^pr]^. That these
large numbers of quinqueremes were not all quinqueremes is now almost a
commonplace. Other wars apart, we know that in this one both sides had
triremes 59 and pentekontors, 60 and the Carthaginians quadriremes. 61 The same
usage of quinqueremis is not infrequently found in the third decade of Livy, no
doubt taken from Polybius ; and Livy sometimes supplies a sort of proof that
quiiiqiiercmes do not always mean quinqueremes. 62 The real question, of

85 Mylae 50, Sardinia 20 to 40, Tyndaris 18, passim) ; and the explanation must be that

Ecnomus 100, Hermaea something over 114, Polyxenidas' battle fleet of 100 ships was ac-

say 130, Aegates Ins. 120 : or about 438 to 458 companied by some 100 light craft, which Livy

all told. Polybius adds another 30 or so for has not given. As we fortunately know that

the supposed battle before Mylae. 10 years earlier Philip V. had fought in these

56 Boodes takes 17, Tyndaris 9, Ecnomus 24, waters with a fleet containing 150 lembi to 53
first storm (with Hermaea) 170, second storm cataphraets, we can sec that Appian is probably
27. year of Drepana 223 = 470. Add some 10 right, especially as Polyxenidas was engaging
for Mylae, and an unknown loss at the Aegates an enemy 151 strong : and for once we com-
Insulae. Polybius gets his figure by counting pletely justify the larger number.

the Hermaea prizes again, and adding another : " Polyb. 1, 53, 9, o't ■npoir\t'iv tldur pivot

123 for the second storm. — Refitted prizes make \ifj.Qot.

the Roman and Carthaginian totals overlap to 59 Polyb. 1, 20; Duilius' column.

some extent, perhaps 70 to 80 ships. ,; Polyb. 1, 20 ; 1, 73.

57 Perhaps I may give one instance of what I 6; Polyb. 1, 47.

mean. Battle of Corycus, 191 b.< . ; Livy and Ba For instance, Marcellus' fleet before

Appian agree exactly as to the Roman fleet, but Syracuse is 60 quinqueremes Polyb. S, 4 (6) =

Livy gives Polyxenidas 100 ships (70 cataphraets, Liv. 24, 34 : but he has a quadrirem< I.

SOapertae), while Appian(6'yr. 22) gives him 200. 30) and 2 triremes and 3 smaller crafl Liv. 26,

Now Appian, who himself wrote on the Roman 39). (I do not say that the smaller craft are

navy, does not, as a rule, throw naval numbers reckoned in the 60 ) Again, Liv. 21, 1H and 50,

about anyhow; (for the proof of this see the praetor M. Aemilius cuts off and captures 7

Kromayer's article in f'hilologus before cited, Carthaginian quinqueremes, with 1,700 milites

60 THE FLEETS OF THE FIRST PUNIC WAR.

course, is Polybius' credit ; and I think we can go a little nearer than the
mere assertion, no doubt partly true, that he used irevrrjprj^ simply for
' warship,' as some writers use Tpirjprjs.

There can be no doubt that, for the Roman navy of the third and second
centuries B.C., at any rate, the quinquereme was the standard warship, quite
apart from the first Punic war. If Rome engaged to aid a foreign power, it-
was with quinqueremes. 63 If a victory was to be announced, a quinquereme
was sent. 64 Envoys and commissioners always sailed in quinqueremes, usually
one apiece. 65 It was the typical Roman ship ; and after 260 B.C. was very
likely almost the only type of ship built in Rome itself, 66 seeing that the
treaty cities supplied open vessels, triremes or lesser, and did not (except
Carthage after 202 B.C.) supply cataphracts.

Fortunately, we do possess one trustworthy piece of evidence of the
composition of a Carthaginian fleet in 219 B.C. ; probably a Roman squadron
was very similar. When Hannibal set out on his march, he handed over to his
brother Hasdrubal his ships, consisting of 50 quinqueremes, 2 quadriremes,
and 5 triremes. 67 Polybius rather apologises for being so precise, but says he
took the details from the inscription on bronze, which he had read, left by
Hannibal himself in the temple of Hera Lacinia. This would make the
proportion of quinqueremes in a squadron sometimes as high as 88 per cent. 68
The Afcadian, mediterraneus homo, may be pardoned for talking of a fleet as
a fleet of quinqueremes when in fact 12 per cent, of the number were some-
thing else ; while the philosophic historian would certainly consider the dis-
crepancy supremely unimportant. When Polybius has good authority
before him, Hannibal or an admiral of Rhodes, he gives precise details;
elsewhere it may be that he is satisfied with conveying what he considers to
be a substantially correct impression ; and, after all, he himself had seen a
fleet of the old Roman navy, perhaps the last of its fleets to go into action.
For that navy scarcely survived the destruction of its great antagonist ; and
Rome was content to fight with ships of Greece and Asia until the lex
Gabinia opened a new chapter in the history of the sea.

W. W. Tarn.

nautaeque, i.e. about 242 to a ship ; the roivers exactly that of the 74 under Nelson; the

on a quinquereme were more than that, according quadrireme and trireme corresponded to the

to 1 orybius. smaller ships of the line of 60 or fewer guns, the

3a Liv. 26, 24, the treaty with Aetolia. That 'light craft' to frigates and brigs, while the

quinqueremes were in fact s.-nt appears frtm heptereis and dekereis of the Hellenistic powers

Liv. 27, 32, where on 15 of the ships Sulpicius took the place of the ships of 110 and 120 guns

terries 4,000 troops over the Gulf of Corinth, built by France and Spain. Though both

giving an average of 266 on a ship, which quinquereme and trireme fought in the line, the

Kromayer says is the highest to be found. fact that Livy classes triremes among ships

i olyb. 10, 19 = Liv. 26, 51. minoris foriuae, as opposed to the quinqueremes,

15 Polyb. 15, 2; Liv. 28, 17 ; 29, 9 ; 30, 25; &c, majoris formae (37, 23; 36, 41), shews

30, 26 ; .31, 11. some well-marked distinction between the two

I think there is no instance of any other other than mere size ; no doubt the line of

type being built. division is between the galleys with little oars

"olyb. 3, 33 = Liv. 21, 22. rowed by one man and those with great oars

It I may venture on one modern parallel, rowed by several men, a distinction which to

the place of the quiuquereme at this time was a spectator would be most conspicuous.

INSCRIPTIONS FROM THE CYZIOUS DISTRICT, 190G.

[Routes followed : («) Pandemia, Mihallitch, Kermasti, Kavakly,
Susurlu, Eski Manyas, Yeni Manyas, Alexa, Pandemia; (b) Pandemia, Erdek,
Karabogha, Gunen, Pomak Keui, Hautcha, Pandemia; (c) Soma, Balukiser,
Mudania, Brusa.]

1. Pandemia, in private possession: stele 0'40x0"25 m., top broken,
with relief : —

(1.) Worshipper with sheep; altar; (r.) Apollo standing, with kithara in
left band, patera in right. Below (letters , 015) :

HAIOAQPOZ 'HXioSw/oo?

Al"10 A Aft Nl ' AiroWmvi.

The relief is of a type very commonly found in the district ; cf. J.H.S.
xxiii. 87 (39), xxiv. 20 (1), xxv. 58 (13).

2. Kermasti, at the Konak. Two fragments of white marble epistyle,
consisting of dentils, frieze of bucrania, and double architrave : both frag-
ments have been broken across. The architrave is 0*12 m. deep, the whole
epistyle 030. The architrave is inscribed in letters "02 high with apices and
broken crossbars in H, E : —

(a) 0'64 long, frieze missing :

HTOnOAEITHNTHNArAOHNTH
EKTHNIAinNEYIXI-IMnh
(6) 0-60 long:

TONNAONAYTHIKATEIKEYAir

3AEnSTYXHNKAI

3PcpYPonnAi-iz

The whole therefore runs :

MetX]>/T07roXetT<wi> Trjv ayadi]V rrj[<; 7r]o\e&>9 Ti^>/z; /cai tov vaov avTrjs

KaT€<TKevaae[v \
6K twv ihlcov Eva^ficov 7rop^)U/?07ra»\>;?.

62 F. W. HASLUCK

These fragments probably came from an excavation on the site of
Miletopolis at the fifth kilometre stone from Kermasti on the road to
Mihallitch, where I have seen at various times a quantity of Roman
architectural detail including green marble shafts, white marble twisted
columns, elaborate circular ceiling panels, and various fragments of archi-
traves, etc. The site was apparently plundered for the building of the
mosque at Kavakly, where there is a corresponding green marble column
A Tyche, not specially characterized as a City Tyche, and never in a temple,
occurs on Miletopolitan coins of Crispina, J. Domna, Gordian III. (medallion)
and Otacilia (Mionnet, ii. Nos. 363, 364, 368, 371;.

3. Ibid. : Yellowish marble slab, broken top and bottom, 0'35 m. (left
edge) — 029 (right edge) high, 020 broad, '085 thick : dowel holes in both
edges: letters '01 — "0075 high. From two squeezes 1 : —

I 1^

IkA bEYfE eAy-TONEYPoI

. A PTYPFIoT 1 A EYTTpcIl iropo£r.Nc
lAPTYPLP^ ?irroW|NOYEr*AiP^

^c.NH2KP\^lr.oNCi/wETRYKCk:

,-_- w\Ai I NMLT^^oE

-YXHNno |=r c , Q ] A ^ on?r <n

OO'^OIA^' ' "* {IaW^YAAIU
w/Xir^p- \WrtBoYA£YoY>^ oNr ,

tA|AElA£EXoY ^^No,,\|AiKLI

\0~~ N.\|.n.|<E H^ENC A-rvN<j>PoH
s p F - HN r r A,, El^'^lH, 'rcPYl T,

Tp/ 5TE:A||<AlA^ P ,A,ni-EY E

iN^r^Ao^ aa r rpo>HAONHh
lAOlZ'C'YNcri nrc^KY^ lTcut|or.
XopoY^A M Y N o , V / \iP oh "" ,'oraEfcoY
-Tr-F>FI*A r ^PA^lAA Y o Y

•\|<,A »./m "UPsOlrHpArppotAEXp^
O I N O £ r 1 N OY *P A • ^ * YM 4>EP<?N

Y^H^O^ n NOT HAOTT El

-r^"AxF A CE:ioM OAc '

1 The better is too broken for reproduction, whence discrepancies between the facsimile and
transliteration.

INSCRIPTIONS FROM THE CYZICUS DISTRICT, 1906.

(p]c\oi<; (3oij6ei.
6]vp.ov Kpdret.
a]8iKa (pevye.
fiaprupei oaia.
5 rf\hovt]<i Kpt'iTGi.

ir\povoiav (rtp.a ?).

a]8c/ca> /i?; XP&.

(p]iXi'av dydira.
10 ci~\vaihela$ («7r)c^of.

86£ai> Scoo/ce.

dperrjv eiralvei.

irpdacre Bitcaia.

Ti\/J.i]v a7roSo?.
15 cpi\oi<i evvoei.

i]X0pov<; dfivvov.

<Tvyyei>€is cia/cei

K^aKLas uTre^ou.

fj.ia]oivo<; ytvov.
20 t]« i&ia (f)v\aaae.

(piXwi yapi^ov.

vfiptp, p.€t<T€t.

eixpripios ylvov.

oi]K€Ta<; eA.ee/.
25 ... o<9 . .

7rt(p)a<; e . . .

irdcriv (pi\o(f)[p6i'ei.

yvvaiKos dpx\e-

aavrbv eu 7rol[et 1

evirpocn'iyopos yivo[y. 5

diTOKplvov ey fcatpco[t.

irovei fier €v/cX€i[a<;.

dpLaprLav p.era[v6ec.

6(f>6aXp,ov Kpdre^t,.

(piXcav (pvXaaae. 10

(SovXevov xp6va>[i.

7rp[a]crae avv vo/jlcoi.

t[o Bik Qatov vep,e.

opovoiav cia/cei.

/j,]t]0ei'b[^] /caTa(pp6i>[ei. 15

ciTropprjTa Kpinrre.

t]o Kparov/J, (pofiov.

yjpovtoi ircareve.

vale 7T/30? rjSovijv.

TTpOGKVVei TO 0€to[v 20

/caipop. 7rpoa8e)(o[v.

eydpav SiaXvov.

eVi poop-ijL {firf) Kavyoi.

yrjpa<; TrpoaSe^ov.

XP<*> T< ^ i o-vfi<pepovT[i. 25

ev^fpLtav dcrfc[ei.

■tyevhos al<ryyvo\y.

dirkyOeiav <f)6vye.

iriarevco[v] firj a . . .

irXovrei a . . . .30

6pboX6y[et . . .

The date of this curious collection of aphorisms seems from the ortho-
graphy and lettering to be about 300 B.C. Its purpose will probably never
be known unless the preamble of it or a similar inscription comes to
light.

The following copies of inscriptions were courteously communicated to
me by Mr. D. A. Renjiperis of the Regie.

4. Alpat Keui

THTONAPETHIE
KAIEYNOIAITHIE

arpa\jrjyhv aperr}? e[ve-

ko\ /cal €vi'ot'a<; tt]<; e[l<; eavrov ?

64 R W. HASLUCK

5. Melde (the site of Miletopolis) : stone with right edge broken.

ERRATAOVAE

ITINIARMINIA in<i> Arm(e)nia

BENDOMITIO su]b (C)n. Domitio [Corbulon- ?

EI EG i e (l)eg[atus ? Caesar- ?

: SAVGPROPR is Aug: pro pr[aetore ?

PFSOAASPRO

OHONOR • C • AVS honoifis] cans[a ?

AHPAnAPAXEIMAI iv % A&\hr,pq 7rapa X ec/xda[a^
nONAlON

The Latin is evidently ignorantly copied, but the mention of Armenia
suggests that the inscription refers to a legate of Cn. Domitius Corbulo,
possibly Ummidius Quadratus (Dessau Prosop. Bom. 600), whose family later
held office in Asia; but conjecture is unprofitable till a better copy of the
inscription is to hand.

6. Melde: letters with apices, A, E with disconnected cross-strokes: —

/ TITOCCJ)/ Tito? < t>\[aovio<;

OYIOYAC Titov (v)lov 'A«r[/c\^7r-

PEINAAT uiSrjs Kv]pelva aTrfoTeXecra?

TEIMACTHNCEBAC rei/xas rojv %e/3ao-[T(t)v iv tg>

AHTOnOAEITHNlEP Mei]\r)T07ro\eira)v iep<p

XHCAC teal yup,paaiap] X r]aa<i ?

7. Melde :

OAH[MO]l 6 8i][fto]<;

NAlONnOMnHIONTNAO T]vdlov Uo/xinjiov Tva(t)o[v

ONMATNONAYTOKPATOPA v [fi v Mdyvov, avroKpdropa

3TPITONIHTHPAKAIEYEP T ]o rpirov, amrfjpa kul edep-

ETHNTOYTEAHMOYKAI y]eTr)v toO re Btjfiov teal

THIASI AinAZHSEflO tt}? Walas Trd<Ti)$, eVo-

THNrHITEKAl©AAAS 7t]tt]v y*}<? re /cat daXdcr-

HIAPETHIENEKAKAI &]*]<;> aperf}? eveica zeal

NOIAZEISEYAYTON ev]i>oia<; el? e<v>avToi>.

Other honorary inscriptions of Pompey in Asia have been found at
Thymbra 2 and in Mytilene. 3 The present dates from the passing of the
Manilian law (66 B.C.), but Pompey is not known to have been in this
district at the time, though acorfjpa ical evepyeTtjv r ov 8 ij jjlov evvoias eve/ca
669 eavrov seem to imply personal relations. Miletopolis makes its first

- C.I.G. 3608 = Dorpfeld, Troja, 58. :i I.G. 163-5.

INSCRIPTIONS FROM THE CYZICUS DISTRICT, 1906. 65

appearance in history as the scene of Fimbria's victory over Mithradates in
85 B.C. 4

The character of the above group of inscriptions, especially the early (3)
and the important (7) makes the attribution to Cyzicus of the long series of
inscriptions from Ulubad more than ever problematical.

8. Yali Chiftlik : copy of M. Alphonse Serafimoff.

ATAOHITYXHI 'A 7 a6T/t Ti/^t.

AYPCW4>PONlOC Avp. 1a><pp6vio<:

OENEriHKOH 0e$ i-rrriKOtp

AlONYCNKEBPhN biovvom Ke/3pr)v[{<p ?

EYXAPICTHPION €v X apiaT7Jptov.

Dedications to Dionysus are scarce in the district (Lebas 1100. Mihallitch,
Ath. Mitth.'ix. 17 (3), Panderma, J.H.S. xxv. 57 (7), Gunen) : Bromios and
Mystae are mentioned in an inscription of Beychiftlik ( = Yali-chif tlik) i?.6'.if.
xxv. 874 (20) = 5«/o^ai/»79 i. 330 (12).

9. Yeni Manyas, in the street : marble block 077 in. X 064 with relief
of wreath in sunk panel ; below, inscription, 0"29 deep, in letters '02 high,
much worn and defaced.

OY . . . AHO

OMO0

THAIOSTHZ

STE(j)ANOYZ€ STO
lANTAAlAIH NH
I . . (J)ANOOIN IA

HZANTAAIAIf in

IITE(j>ANOY N BO

NO0ETHZ inZ

The honorary character of the monument is shewn by the relief and the
frequent occurrence of aTecpavos and alSicos in the mangled inscription.

Tchakyrdja : —

10. In private house: fragment of slab with sunk panel; on edge, in
letters of late form "03 m. high :

ICANTIOXOY ®4<r]v;'AvTi6xov.

4 O.I.G. 6855. Cf. Memnou 34, Frontin. iii. 17. 5. Oros. vi. 2. 10.
H.S. — VOL. XXVU. F

66 F. W. HASLUCK

11. Step of school-house: altar-shaped stone, 1 - 10 m. broad, 0*50 high.
Along the upper profile (letters "035) :

DCECTHTICT <- 0-45 -> AI-MATIEI

v7T€v6vv]o<i €<tt(o r(f])^ r[vp, {3 copulas ey/c]\i)fia,TC el [ Se T£? etc.

12. In private house : marble stele, 075 X 0'48 m., with relief of (from
left) four worshippers, sacrifice of bull, large plain altar. Below (letters "02'
high) :

MEAEATPOEKAIGEOZENOS MeXeaypo? fcal Seogevcx;

KAIMENANAPOZOinPHTOMAXOY kcu MevavSpos oi UpoTo/jbdxov

AIIBPONTAiniEYXHN Ad Bpovratwi evytfv.

The stone is said to be from a site between Tchakyrdja and Hadji Paon.
The dedication to Zeus Brontaios is interesting in connexion with the
autonomous coin-types of Poemanenum, 5 Obv. Zeus head ; Rev. Fulmen.
Zeus Hypsistos Brontaios is mentioned in an inscription from Mihallitch
(Lebas 1099 = Mori. Fig. PI. 133. 2 and p. 115=Bev. Philol. i. S8 = Ath. Mitth..
iv. 21. Tchinili Kiosk Gated. Sculp. 126).

Poroak Keui (left bank of Aesepus, half an hour below Gunen) : —

13. In the street: marble block, 0*91 m. x 0485, '05 thick, letters -045 v
much worn :

l AlOllllEIMOinAPANrEAAinAII
DIITHNAEYKEANKOriTOYIlN
HNAPINKAI . T. \TINAAnr

nn in iiiu iioiitoytx

AIOY
FAnol I Tf T

N ITI

'\o\vXios (JIp)eip,o<; irapavyeXXi iraai
r]ot9 rtjv XevKe(a)v kotttovctiv [Sovrao ?
B]r)v(d)piv Kal . .

The stone is said to have come from a site on the left bank of the
Aesepus opposite the hot springs of Gunen.

XevKea-(l) Xevicaia, the white poplar, (2) axotvos (Hesych.), rope: an
announcement engraved on stone can only refer to the former. The white
poplar was associated with Zeus and Herakles (see Frazer on Paus. v. 5. 5,
Boetticher, Baumkultus, p. 441 sqq.) and the tree referred to may have been
one of special sanctity : it is noteworthy that Julius Primus does not forbid
the cutting, but makes a tariff-charge as if cutting was habitual. 6

5 J.If.S. xxvi. 23 ff. known from a Conn inscription (B.C.ff. xxiii.

8 A festival called AevS^oiccnriov ttjs "Upas is 208).

INSCRIPTIONS FROM THE CYZICUS DISTRICT, 1906. 67

14. In a garden : stele, 0*G4 x 33, with relief of man on couch, woman
seated, table, and two slaves flanking the group : below (letters, *015) :

APTEMEI 'Aprefiel^

£NTITTATPOY ! 'AvriTrdrpou

XfilPE

X aipt

14a. The inscription, republished with a commentary by Dr. Wilhelm in
Bcitrdge zur Alien Gcschichte (v. (1905), pp. 2 ( J3-302), is still to be seen in the
ihancel of the church of S. Nicholas at Chavutzi (Kiepert's Hautcha). The
stone measures 0"85 x 0'G3 m., the left edge being entire : my copy agrees with
Linmios', except that he omitted to note two upright strokes remaining from
a line above his first, and the possibly significant fact that his first line
ZTPA . . is set back from the left edge of the stone, as if it had formed the
heading CETpa[ri]yoi ?). Noticeable peculiarities in the lettering are (a) thin-
ness of strokes throughout; (b) tendency to cross the ends of coincident
diagonal linos (JS, X, etc.), which gives somewhat the effect of apices;
(c) variation in form of letters: thus p in arpa . . is written R, elsewhere P,
TT varies between F 1 and FT ; (d) variation in size of letters : they are
normally (11. 2, 3, 9, 11, 13, 14, 15) -025 high, but rise to '03 in 11 4-8, 10, to
0325 in I. 1 and to -045 in I. 12. These irregularities suggest that the
names were added to the list year by year.

15. Tchatal-Aghil, near Brusa, church of Theologos; slab 0*82 by
0*66 : in tabula ansata - 32 x 0T8, letters - 025 :

EYBOYAE EvfiovXe

0EO(f)iAOY QeocplXov

P E X al ]P € -

Below, relief of Herm in niche.

This stone was seen by Stephan Gerlach (1576) ' half-way between

Ulubad and Brusa.' "

F. W. Hasluck.

7 Tiirckisches Tagebwh ( Frankfurt, 1674), p. 257.

r 2

TERRACOTTAS FROM BOEOTIA AND CRETE.

It is proposed in this paper to deal shortly with five terracottas in the
possession of the writer, which seem of sufficient interest to be illustrated.
The first three are from Boeotia, the last two from Crete.

As is well known, the most primitive Greek standing figures in terra-
cotta frequently take one of two forms — the columnar form, derived probably
from the tree trunk, and the flat broad form (aavk), taken apparently from
the shape of a board of wood. These forms confirm the literary evidence that
the most primitive statues were made of wood. The first three terracottas
illustrate the latter type ; in the second of them the boardlike form has been
adapted by a singular device to a seated figure.

Fig. 1. Primitive standing figure from Boeotia: height 024 m., greatest
breadth 0"84 m., thickness 0*012 m. With the exception of the head and feet,
the figure is absolutely flat. The arms consist of fin-like excrescences ; the feet
are perhaps the best rendered part of the figure ; otherwise the forms of the
body are not expressed at all. The edge of the drapery, nowhere else indi-
cated, is sharply defined above the feet.

More attention has been paid to the head. The primitive artist has
grasped the fact that the eyes are the most important feature of the face, and
has represented them by large disks of clay surrounded by a deep rim. The
nose is long, narrow, and fiat; the mouth is not indicated at all. The hair is
formed from one long string of clay, which is bound over the forehead and
crosses behind the head ; thence it falls in a long lock, covered with a number
of small indentations, on either side of the face.

The artist clearly worked with the simplest of tools, apparently pinching
the clay for the most part with his own hands— this is evidently the way in
which the nose has been formed — and using also a pointed instrument for the
feet, hands, and hair. There are traces of a slip of white material over the
whole figure.

Such o-az^Vlike figures are common, especially in Boeotia. 1 I can find
no close parallel to the treatment of the hair and features.

Fig. 2. Primitive seated figure from Boeotia : height 0145 m. The whole
workmanship of this figure, when compared with the last, seems to show that

? — •

1 Cp. Winter, Die ant i ken Terrakottai, iii. pp. 4, 5, 9, 3].

TERRACOTTAS FROM BOEOTIA AND CRETE.

it is from the same hand. It is an interesting variety of the ordinary primitive
seated type ; instead of the usual solid throne or seat, the figure is sustained
by a support behind, rather in the style of a modern photograph frame. The

Fig. 1.

features of the face are almost identical with those of the first figure; it has
the same owl-like eyes and bird-like nose, still further exaggerated. The hair

EDW. S. FORSTER

is rendered in the same way, but not decorated in front with indentations.
The feet and lower edge of the drapery are not represented.

The figure holds a smaller figure clasped to the breast with the left arm,

Fig. 2.

which is far larger than the right arm. The features of the infant are those
of the larger figure on a small scale.

It is interesting that the partiality for the boardlike form is so strong
that it has been adapted to the seated figure. The seated female figure hold-
ing an infant is one which is common among terracottas of every period.

TERRACOTTAS FROM BOEOTIA AND CRETE.

This type of propped seated figure has been found in several Greek sites,
•&.g. at the Argive Heraeum 2 and Tanagra. 3 The only parallel to such a

Fig. 3.

figure carrying a child is to be found in a terracotta from Boeotia at Berlin : 4
this is of a much later style.

- Waldstein, Excavation of the Heraeum, PL
VIII. Wis. 11.

:; Athens, National Museum, Xo. 833.
4 Berlin, Antiquarium, Xo. S34S.

EDW. S. FORSTER

Fig. 3. Standing figure from Boeotia : height 0'235 m., average thickness
0011 m. Here the form of the body is exactly similar to that of No. 1, except
that the arms are longer and the feet are not indicated. The figure is covered
with a slip of white material, on which are faint traces of red paint run-
ning perpendicularly down the centre of the body and horizontally at each
side. The head, which is damaged, shows a very considerable advance. The
eyes, though not exactly in the right plane, are carefully rendered, and the
cheeks and chin are well modelled. The head is surmounted by a high polos
coloured red ; the hair is scarcely indicated.

Fig. 4.

By analogy from sculpture in stone, the figure seems to belong to the late
sixth century. It is interesting to see that the conservative instinct of the
Greek prefers the flat, shapeless form of the body at a time when art is suffici-
ently advanced to enable the far more difficult features of the face to be
represented with some success. In this respect the rendering of the human
form in terracotta differs from that in stone, where perfection begins with the
feet and finally reaches the head. In works of sculpture the artist naturally

TERRACOTTAS FROM BOEOTIA AND CRETE.

tried to excel in every part of the work ; in terracottas, religious conservatism
demanded the form to which it was accustomed, and it was only in the head
that the artist ventured to use his growing skill and knowledge.

The two Cretan figures, which were found together in a tomb near Retimo,
are both equestrian, and represent uncommon types.

Fig. 4. Group of two horses and driver : height 012 m., extreme breadth
01 m. This group is made of dark, coarse clay, baked very hard and covered
with a slip of white material. It represents a man mounted on a chariot;
but, owing to the exigencies of the material, horses, man, and chariot
are all moulded together. The chariot wheels, which are solid, appear on
either side of the back legs of the horses. The structure and attachment
of the chariot are not shown, and its presence is only indicated by the wheels.

Fig. 5.

Tt represents no doubt a war or racing chariot, consisting of wheels with a
cross-bar supporting a platform on which the driver stood ; the convenience of
a war-chariot which could be easily mounted or dismounted from is obvious,
and the racing-car preserved the form of the war-chariot.

Chief attention has been paid to the horses. Their heads are carefully
modelled, particularly the ears and crest ; their manes are indicated with black
paint, of which considerable traces remain. The head of one is raised, giving
a pleasing variety. The legs and back are only roughly blocked out.

Less attention has been paid to the charioteer. His head is rudely
sketched, the large curved nose being the only marked feature ; the forms of
the body are not rendered at all. On the left side there are indications that
the arms stretched along the horses' backs. This fact and the position of the

74 TERRACOTTAS FROM BOEOTIA AND CRETE.

horses' front legs, which are planted firmly on the ground before them, seem
to show that the group represents a charioteer reining in his horses.

Fig. 5. Horseman carrying faggots : height 0T45 m., extreme width 0T15 m.
'This group is of the same material as the last figure, and was apparently also
covered with a slip of white paint, of which few traces remain. Here the horse
is subsidiary, and the chief attention has been paid to the rider. The horse is
of the form found on many Greek sites, and little attempt is made at modelling ;
the legs and tail are thick, the head small. The faggots consist of spikes of
clay, built up on the top of one another. The rider sits astride on the top, his
legs projecting in front and his hands on either side of the horse's mane. His
face is clearly intended to be of a comic character. He wears a pointed beard ;
the lips are thick and the mouth large. The ears are shapeless excrescences;
the nose is large, and the eyes are formed of disks attached to either side of it.
The head rises to a peak' at the back. In short, the figure is of that grotesque
character which occasionally appears in Greek art of every period, from the
early sculptures of the Athenian Acropolis down to the late Hellenistic grot-
esque terracottas of Asia Minor.

As is the case with the other Cretan terracottas, these equestrian figures
find their closest parallel among the terracottas of Cyprus. 6 They seem to
show that in Crete too is to be found something of the charm and naivetd of
primitive Greek art of the mainland.

Edw. S. Forster.

5 Winter, op. cit. p. 15.

ON THE 'LIST OF THALASSOCRACIES ' IN EUSEBIUS.

All students of Eusebius will feel grateful to Mr. J. L. Myres for

his attempt, in the last volume of this Journal 1 to discover the original
text underlying the list of thalassocracies, preserved in the Chronica of
Eusebius, and to reassert its value as historical evidence. The problem
that Mr. Myres has set himself is rendered difficult not only by the
general obscurity in which the sources of early Greek history are shrouded
and by our almost total ignorance of the history of many of the thalassocrats
daring the period assigned to them in the list, but by the complicated
questions of textual criticism which surround the Chronica, and which this
problem raises in a particularly aggravated form. While not venturing to
follow Mr. Myres through the wealth of historical learning, which he has
brought to bear upon the subject, I have thought that I might be able to
contribute something by bringing my own studies in the Chronica into
relation to the general question.

From this point of view, §3 in Mr. Myres's article, which deals with the
text and its use by Eusebius, is the most important. It is to be regretted
that Mr. Myres, whose article shows an acquaintance with the essays on
the Chronica contained in Von Gutschmid's Klcinc Schriften (1889), and with
Schone's Weltchronik dcs Eus, bius (1900), has not devoted a little space to
the general critical problem of the Chronica, the more so, as he sometimes
drops into a phraseology not consistent with the views he quotes from
Schone. For instance, he frequently refers to the Armenian version of the
Chronici Canones as Eusebius, and to the Latin version as Jerome, as though
the Armenian version were identical with the original Greek in a way
in which Jerome's version is not. In one passage, p. 92 21 , he distinguishes
between the Canon and the Chronicon, as though the Chronicon were
identical with the Chronograph ia, instead of being the title of the whole
work. A brief description of the Chronica and of recent critical opinion
in relation to it may therefore be useful with a view to removing confusion.

The Chronica of Eusebius was a Greek work in two books, called
respectively the Chronographia and the Chronici canones. The Chronographia
is a chronological treatise, consisting largely of excerpts from previous writers,
and is preserved in an Armenian version. The Chronici canones or canons

1 For convenience of reference, Mr. Myres'a table is rebate,! on p. 76 of this article.

J. K. FOTHERINGHAM

]

Diodorus

quoted by |

Eusebius,

Chroni-

con I.
x 168 Mai,

p. 225
Schoene.

Syncellus
— ] shows
that the
name is
given
but the '
numeral
omitted.

Eusebius'
Canon,
years
from
Abraham,
and dura-
tion of
>ea-power. <

Jerome's
Canon
years
from

Abraham,

xnd dura-
tion of

ea-power.

Euse-
bius'
dates,
reduced
do years

B.C.

Jerome's

dates,
reduced to
years B.C.

Winck-
ler's

dates,
years

B.C.

Bates and
intervals

now
proposed.

1184

[Caves]

128 1S8

848

842

1168

1174

1056 1046

Lydi (qui et

[-3

[-]

118

Maeones)

928

960

1088

1056

964 954

Pelasgi

[-1

{ 1009)
\ 1050 \

49. 90. 95

1 1007)

\ 966 V

85
879 869

III

Thiaces

(.1055 J
19

l 961 J
92. 41. 46

c > ©

3 i-i

1101

915

757

S00 790

Rhodii

1113

1123

903

893

734

767 757

Phryges

25 or 6

25 (20)
1152

29
864

25
709

25
742 732

Cyprii

[-]

a go

23 (32)
1180

28
836

33
676

33 23
709

VII

Phoenices

[-]

5 C^

~<~

1233

783

631

664

VIII

Aegyptii

—

[-]

[-]
1268

748

26
605

60
604

Milesii

—

1286

1296

730

720

587

[Cares]

—

[-]

1345

671

587

Lesbii

—

[96]
1441

575

578

XII

Phocaeenses

44
1486

45
530

44
534

XIII

Samii

—

[-]

XVI

1503

17
513

17
517

XIV

Lacedaenionii

515

Naxii

1505

In J

505

XVI

Eretrienses

1531

1508

485

508

490

XVII

Aeginetae

10
480

1 Reprinted from Mr. Myrcs's article, Vol. XXVI. p. 88.)

ON THE 'LIST OF THALASSOCRACTES ' IN EUSEBIUS. 77

are a chronological table, extending from the birth of Abraham to the
twentieth year of Constantine, and are preserved not only in the Armenian
version already mentioned, but in the Latin version of Jerome, while a
few entries are reproduced in a Syriac epitome. The original Greek of
both books is lost, but can be restored in large measure from quotations and
parallel passages in Syncellus and other writers. Schone in his Weltchronil;
des Eusebius has argued that of the two versions of the Chronici canones
the Armenian has completely transformed the original arrangement of
columns and spaces, while Jerome adheres line for line to the arrangement of
his original. I have endeavoured in my introduction to the Bodleian
Manuscript of Jeromes Eusebius (1905) to show that Jerome adheres not
only line for line but page for page to his original so far as the part of the
canons down to the year of Abraham 1504 (512 B.C.) is concerned, after which
date he would appear to preserve only the general arrangement, not the
exact lineation and pagination of Eusebius. Jerome does not attempt to
correct Eusebius, but has made numerous insertions, relating mainly to
Roman history. The Armenian translator on the other hand neither amends
nor adds to his original, but often omits events, apparently by oversight.
Schone holds that the Armenian version is made from an earlier, and
Jerome's from a later and revised edition of the Chronici canones of Eusebius.
His reasons for this are stated in his Weltchronik etc. pp. 260-7. In
my introduction quoted above I have left this an open question, but further
study has convinced me that the differences on which Schone has based
his case are with one exception of the same kind as the differences between
those manuscripts of Jerome which retain the original arrangement and those
which, like the London manuscript, have substituted a different arrangement,
drawn up with no great care. 2 The single exception is the inclusion in
Jerome of a column of Mycenaean kings, which is absent from the Armenian
Canons, though such a list is found in the Chronographia. It is impossible to
base a theory of a separate edition on this one instance, since the Armenian
translator, who, as we have seen, transformed the arrangement of the book,
might easily have omitted a column by accident. 3 It would therefore appear
that the differences in arrangement, like the differences in text, between
the two versions, are for the most part the result of errors on the part
of one or other translator or of some copyist. The kind of error that is
most common in manuscripts of the Chronici canones is for entries of event*,
while retaining their position in relation to other entries of events in
the same column, to be shifted upwards or downwards, so as to stand against

2 Am ng the instances cited by Schone are (1900), i. ]85.
the differences in the dates of Roman bishops, 3 The Armenian version also omits the

on which much has been written. I am column of Median kings, although this column

convinced that the peculiar dates of the is essential both to the chronological system of

Armenian version are simply due to Bcribal the work and to the arrangement of columns

errors, and ilo not »o back to Eusebius himself. and spaces, as preserved in Jerome. The

See Mr. C. II. Turner's article on the Early omission of the Mycenaean co'umn would not in

Episcopal Lists, Journal of Theol itself dislocate any of the other columns.

78 J. K. FOTHERINGHAM

different years, and, where there are two columns of events, to change
their position in relation to entries of events in the opposite column.
It is therefore important in considering suggested corrections of the text to
observe how far they involve alterations in the order of entries occurring
in the same column or space.

The thalassocracies occur in two places in the Chronica, (1) in the
Chronographia, represented by the Armenian version only, (2) in the
Chronici canones, represented by the Armenian version, by Jerome, and,
according to Von Gutschmid, whom we shall see reason for following, by
Syncellus. There is no reason for suspecting either translator of Eusebius
of having attempted to do anything else than render the text as he found
it. In the case of Jerome the amanuensis would, as far as 512 B.C., simply
keep each entry in the place where he found it in the Greek. Syncellus
might, consistently with the principle of his work, introduce material from
some other source, and in two instances he inserts alternative figures which
may have been obtained elsewhere, but he does not appear in the present
instance to have adopted any date from outside Eusebius. But while there
appears to have been no attempt to improve on the figures given by
Eusebius, it is, as Mr. Myres has pointed out, by no means clear that the
dates for the thalassocracies given in the Chronici canones were calculated
by Eusebius from the list given by Diodorus which appears in the Chrono-
graphia. Eusebius may have somewhere found the date of each thalasso-
cracy already correlated to the dates of other events which appear in the
Chronici canones, and may have placed his entries accordingly. We are
therefore faced with three possibilities. We may have in Eusebius a single
scheme of thalassocracies derived from Diodorus, or we may have two
separate schemes of which one only is derived from Diodorus, or we may
have, as Mr. Myres supposes, one scheme drawn from Diodorus, and a chaos
of dates not calculated on any fixed principle.

Before examining the dates in detail it may be well to see how far
Mr. Myres's table accurately represents the evidence before us. Columns
A, B, and C, giving the order of the thalassocracies and the length of
each as recorded in the Armenian version of the Chronographia are
correct. But in column D, giving the figures preserved in Syncellus, a
few errors may be noticed. Thus Syncellus actually gives ninety-two years
as the duration of the Lydian or Maeonian thalassocracy, agreeing with the
Armenian version both of the Chronographia and of the Canons. He makes
no mention, however, of the sixth, seventh, eighth, and thirteenth thalasso-
cracies instead of merely omitting the figures as Mr. Myres states. In
column E, which gives the figures preserved in the Armenian Canons, Mr.
Myres's only mistake seems to be in the case of the Eretrian thalassocracy,
where he has substituted 1505, the number standing opposite the record of
the event in Schone's edition, for 1514, the date pointed out by Schone's
index letter. With column F, as an accurate reproduction of Jerome's figures
given in Sclione, no fault can be found, except that the forty-five years as-
signed to the Phoenicians are to be found in one manuscript only, and are

ON THE 'LIST OF THALASSOCRACTES ' IN EUSEBIUS. 79

rightly regarded by Schone as no part of the genuine text of Jerome. It
will, however, be seen hereafter that different figures are sometimes to be
found in the Bodleian manuscript from those which appear in Schone's
edition. In column G, in which the dates of the Armenian Canons are reduced
to years B.C., the only errors are those which result from the error in column E.
Thus the Lacedaemonian and Naxian thassalocracies should last eleven years
instead of two, and should be dated 502 B.C., instead of 511 B.C., while the
Eretrian thalassocracy should last for seventeen years only. In column H, in
which Jerome's dates are reduced to years B.C., Mr. Myres has made three
small errors in subtraction. Thus the forty-one years of the Thracians should
be fifty-one, the twenty-three years of the Rhodians should be twenty-two, and
the fifty-two years of the Phoenicians should be fifty-three. Here again the
use of a revised text of Jerome might necessitate a few slight modifications.
In column I, giving Winckler's dates, I have discovered no error, while in
column J, where Mr. Myres gives his own dates, there is one trifling error in
addition. Thus the date of the Rhodians, to whom twenty-three years are
attributed, should be 790 or 780, not 800 or 790, and the dates for the Lydians,
Pelasgians, and Thracians should be correspondingly reduced by ten years, and
the interval allotted to the [Carians] increased by ten years. None of these
errors, except those in columns D and E, affects the documentaiy evidence,
but they are all instructive as furnishing an example of the kind of error
to which we are all liable in transferring figures from one setting to another.
It will be observed that Mr. Myres has nowhere copied a figure incorrectly,
but he has once overlooked a figure, once copied a wrong figure, and four
times made a slight error in calculation. We need not hesitate to attribute
similar errors to the ancients.

The figures given in the Armenian Chronographia present little difficulty.
All except two are confirmed either by Syncellus or by the Armenian
Canons. The two exceptions are the thirty-three years of the Cyprians
and the forty-five years of the Phoenicians. Forty-five years are, however,
attributed to the Phoenicians in codex F of Jerome, a manuscript which is
notoriously the result of a deliberate recension. As the figure forty-five does
not belong to Jerome's text and cannot be obtained by simple subtraction, it
is practically certain that it was obtained by a reference to the original Greek.
This renders it probable that F is also following the Greek in attributing
thirty-two years to the Cyprians, who are assigned thirty-three years in the
Armenian Chronographia, and twenty-three years in Jerome. In the text of
F, they are spread over the period 1150-1181, and are thus made to last
thirty-one years. The figure thirty-three attributed to the Cyprians is there-
fore the only doubtful figure in the list; but we still have the lost figures for
the Egyptians, Milesians, Carians, Lesbians, and Samians to make good as
best we can. When we consider the state in which the Armenian Ch
graphia has descended to us, Ave can infer nothing as to the original from the
absence of some of the figures from our existing manuscripts. It is probable
that the right margin of some ancestor of the existing Armenian manuscripts
was torn. As the extant figures of the Armenian Chronographia are with the

80 J. K. FOTHERINGHAM

one exception mentioned confirmed by figures derived from the Canons, it is
probable that the lost figures would also agree with the figures of the
Canons.

Syncellus, as being in the Greek language, has a special value for the
reconstruction of the lost Greek Canons. That his data are in this case
derived from the Canons, not from the Chronographia, is clear from the way
iu which they are introduced, interspersed among historical events, and
generally among those events which stand close to the notices of the
different thalassocracies in the Chronici canones} Mr. Myres urges that
Syncellus must have used the Chronographia for his figures for the Naxians,
because, as he thinks, the Naxians were omitted from the Canons. But this
omission is merely an inference from the date assigned to the Eretrians in
the Armenian Canons, which Mr. Myres took as 1505, but which, as we have
seen, should be 1514. The phraseology in which Eusebius recorded the
various thalassocracies differs widely, and a comparison of the notices in
Syncellus with those in Jerome will show how closely Syncellus followed the
text of his authority. The following table contains the text of all notices of
thalassocracies, common both to Syncellus and to Jerome : —

Syncellus. Jerome.

Au5ol ol /cot Maioves Ijydi mare obtinuerunt.
iQa\a<TaoKp6.TT)<To.v err? 9/3'.

TleXaayol /3' idaXaaaoKpaT^irav Pelasgi mare obtimierunt.
€ttj 7re .

Tpiroi idaXa<T(roKpa.T7)<Tav Tertio mare obtimierunt

&p§nes (Ti) o0'. Thraces ami XVIIII.

0p5/ces iQaXaaa-oKparow. Thraces mare obtinuerunt

Tfraproi iOaXaaaoKparriaai' Quarto mare obtinuerunt

'PoStoi, Kara 5e rivas ire/j.irToi, trr] Ky'. Rhodii ami XXIII.

4>pvyes irffxirroi tda\a<j<joKp6.T7)crav Quinto mare obtinuerunt

€Tij Kt', Kara 8e rivas erri 5-'. Phryges ami XXV.

'EOaKarfiTOKpdr-qaav Aiyivrjrai XVII mare obtinuerunt Aegitietae.

trrj i'. aim XX.

It will be observed that in six instances out of the seven the two agree
in the relative order of the subject and the verb. The double entry of the
Thracian thalassocraoy, triple in Jerome, is particularly striking, and it is
significant that the phraseology of the two entries varies in the same way
in both writers. This makes it clear that the entry was already duplicated
in Eusebius. Such duplicate entries are common : the explanation probably
is that in addition to the date which he had himself calculated by dead-
reckoning, Eusebius found in some source a date already correlated to some
neighbouring date in his table. The only figures peculiar to Syncellus are
the obviously incorrect twelve and seven attributed to the Lacedaemonians
and Eretrians respectively, and the alternative figure six for the Phrygians.
With this last we must compare the alternative ordinal V for the Rhodians.

4 From the position of the thalassocracies in text of Eusebius which he had before him. This
relation to neighbouring events in Syncellus, will be found useful in determining Eusebius'
we get a vague indication of their position in the date for the Aeginetans.

ON THE 'LIST OF TH AL ASSOCRACIES ' IN EUSEBIUS. 81

This may be due, as Mr. Myres suggests, to the inclusion in some lists of the
Cariaus as the first thalassocrats, or it may be explained by the double
reckoning of the Thracians, the immediate predecessors of the Rhodians.

We now come to the figures preserved in the two versions of the Chronici
canones, and are immediately confronted by the extensive omissions in the
notices presented to us. That no such omissions existed in the Greek text
of the Canons is clear from the fact that each thalassocracy is to be found in
at least one of the three documents, Syncellus, the Armenian Canons, and
Jerome. The omission of several notices from the Armenian version need
occasion no surprise, since, as we have seen, omissions are very frequent in
that version. The omissions from Jerome do not admit of such an easy
explanation. Either these notices could not have stood in the same column
with ordinary events in the Greek Eusebius, or there must have been some
motive for passing them over when copying the notices of ordinary events.
A probable explanation is afforded by the difference of inks of which traces
are preserved in a few extant manuscripts. Ordinary notices in Jerome are
in black ink, but in the Bodleian manuscript, a fifth century manuscript of a
fourth century book, all the notices of thalassocracies are in red ink except
that of the Lydians, which is in black, and that of the Cyprians, which is
omitted altogether. The Fleury fragments (S), perhaps also of the fifth
century, have the notice of the Pelasgi in red ink, but the notice of the
Lydians and the second and third notices of the Thracians in black. Else-
where they are defective. The notice of the Pelasgi is also in red ink in N, a
descendant of S, and there is an erased entry in large red letters at this place
in the Valenciennes manuscript (A, seventh century), but there is now no
notice at all of the Pelasgian thalassocracy in that manuscript. Finally, in the
London manuscript (L, tenth century), in which, as we have seen, the general
arrangement of the work has been transformed, the notice of the Lydian
thalassocracy is in red ink, while the notices of the remaining thalasso-
cracies are in black ink, but the first notice of the Thracian, and the notices
of the Rhodian, Cyprian, and Phoenician thalassocracies, are made to stretch
across the columns, whereas ordinary events are confined to a space marked
out for them. We may, therefore, safely conclude that all the thalassocra-
cies were originally entered by Jerome in red ink. It is impossible to say
whether this use of red ink goes back to Eusebius. Jerome seems to assert
in his preface, 5 that the alternation between red ami black columns was
introduced by him to remedy a confusion that had arisen in the Greek
manuscripts. But it does not follow from this that the distinction of colours
was altogether new. Anyhow, the red ink in which the notices of thalasso-
cracies were written 'must indicate something which distinguished them from
other entries in Eusebius, and which prevented them from being copied out
along with the other entries. It is possible that they were entered by Eusebius
in the margin, This would explain their omission by Jerome even after
the place where the arrangement of the work is altered (511 B.C.;. The

5 P. 2, 11. 23 30 (?chone .
H S. — VOL. XXVII. G

82 J. K. FOTHERINGHAM

theory that these notices were inserted by Jerome separately from the other
entries will explain not only the omission of some thalassocracies, but the
displacement of others. We shall see reason for thinking that the Lydian
thalassocracy has been displaced not merely in relation to the columns of
figures (the fila regnorum), but in relation to other entries in the column for
events. The Rhodian thalassocracy has been inserted in the column which
Eusebius usually reserves for sacred history, and the Aeginetan thalassocracy
has been inserted in the place which ought, apparently, to belong to the
Naxian. The use of red ink or whatever feature in Eusebius is represented by
the red ink in Jerome was probably intended to indicate that the thalassocracies
belonged to the chronological framework of the book or at least formed a
chronological system by themselves ; it at all events differentiates them from
the ordinary isolated events that appear in the two columns of events. This
beino- so, we should expect the intervals between the dates assigned to the
thalassocracies to correspond with the recoide 1 durations of the thalasso-
cracies.

But in the Armenian Canons there is not a single instance where the
intervals given in the text exactly agree with the differences between the dates
either of two consecutive or two more distant thalassocracies. In the general
confusion of the chronology of this version, such a discrepancy need not
alarm us. The case with Jerome's version is slightly different. The critical
apparatus now available for the text of this version is somewhat larger than
that possessed by Schone. The dates given in the Bodleian manuscript for the
thalassocracies differ from Scheme's in six instances. Thus the Bodleian
manuscript has 1016 for the first notice and 1054 for the third notice of the
Thracian thalassocracy, 1100 for the Rhodian, 1125 for the Phrygian, 1234 for
the Egyptian, and 1347 for the Lesbian. Of these figures all except the first
are well supported by other manuscripts and are certainly the true text
of Jerome. For the first notice of the Thracian thalassocracy the best
manuscripts other than the Bodleian vary between 1010, 1011, and 1012.
The Bodleian manuscript also omits the Cyprian confederacy, but there can
be no doubt that Schone is right in inserting it and that he has inserted it
against the right date. The Bodleian manuscript further assigns seventeen
instead of eighteen years to the Milesians, but it meets with no support, and
is certainly in error. If we adopt the readings recommended above, we
find two instances in each of which the length of a series of thalassocracies,
obtained by adding their individual durations, agrees with the interval
between the dates assigned for the commencement and the close of the
series. But in order to effect this, we must adopt the figure 32, which, as we
have seen, was probably borrowed by F from the Greek, for the duration of
the Cyprian thalassocracy, and the figure 61 assigned by the Armenian version
to the Carian thalassocracy, the duration of which is not specified by Jerome.
Thus from the beginning of the Rhodian to the end of the Cyprian thalassocracy
the number of years should be 23 + 25 + 32 = 80 = 1180-1100, and from the
beginning of the Milesian to the end of the Carian thalassocracy the number
should be 18 + 61 = 70 = 1347 - 1268. There is therefore reason for regarding

ON THE 'LIST OF TH ALASSOCRACIES ' IN EUSEBIUS 83

1100, 1180, 12GS, 1347, as not merely Jerome's dates, but Eusebius's, on the
assumption that in Eusebius the intervals between the dates agreed with the
specified durations in each case. It is now possible with the assistance of the
figures already verified by the agreement of the Armenian Chronographia
with Syncellus to reconstruct the earlier part of the chronological scheme of
Eusebius as follows : —

Duration

i>;ue in years
of Abraham

Date B •

]. Lydi.

844

1172

II. Pelasgi.

936

1080

III. Thraces.

1021

995

IV. Rhodii.

1100

916

V. Phryges.

1123

893

VI. Cyprii.

■VI

1148

868

VII. Phoenices,

1180

836

VIII. Aegy[.tii.

1225

791

IX. Milesii.

1268

748

X. <

1286

730

XI. Lesbii.

68(?)

1347

669

It will be observed that while the dates given in Jerome and the
Armenian do not agree in a single instance, the dates given above, which
have been made to agree with Jerome in four instances out of seven, also
agree with the Armenian in one instance out of five. Furthermore, except
in the case of Jerome's date for the Pelasgi, this table nowhere implies
that an event should have been shifted by more than ten years either
in the Armenian or in Jerome's version from the date supposed to have
been assigned by Eusebius, and in the majority of cases the implied
shifting is very slight. Jerome in his preface seems to suggest that some
such confusion had already arisen in copies of the Canons, and, as we
have seen, it is an exceedingly common error so long as it does not affect
the order of notices in a column. That a shifting of two years has taken
place at the beginning of the series is manifest. Immediately after the
notice of the Lydian thalassocracy, against the same date 842, we read ' Mycenis
post necem Aegisti Orestes reguavit ami. XV..' while seventeen years later,
against 859,° we read ' Mycenis regnavit Tisamenus films Orestis.' It follows
therefore that either the accession of Tisamenus has been shifted downwards
oi the accession of Orestes, and with it the Lydian thalassocracy, has been
shifted upwards.

The Pelasgic thalassocracy stands in the Armenian Can ms next before the
Peloponnesian invasion of Attica, which is followed by a notice of the Amazon
invasion of Asia; in Jerome it comes much later. Now there are two
notices of Peloponnesian invasions of Attica about this place in Jerome.
but the one that stands against the year 937 is the one that is
followed by the notice of the Amazons. If then the Armenian version has
retained the original order of these notices, Eusebius must have placed the
Pelasgic thalassocracy in or shortly before 937, a date which agrees well with the

■ : So the Bodleian manuscript. The Flcary fragments hav< :

,; 2

8 4 J. K. FOTHERINGHAM

one suggested in the table above. It will be remembered that the change in
the order of events which must here have taken place in Jerome is highly
improbable except on the theory here advanced that the thalassocracies were
added after the other events had been recorded in their proper places.

The erroneous date for the Thracians in Jerome may be explained on t he-
supposition that his scribe entered the event against the figure XVIII or
XXTII instead of XXVIII in the column of Lacedaemonian kin^s, an easv
mistake in a confessedly hasty work. 7 The second and third dates for the
Thracians. to which no numerals are attached, obviously do not belong to the
series of dates under discussion. The date 1125, which is only two years
in error, needs no explanation. Schone's date 1123. derived from the
worthless Bern manuscript, is only right by accident. The date 11-32
would seem to have been shifted a little further than usual, but it should
be observed that F has 1150, and that there is a very long notice under
the year 1112 in Jerome, which may have made it difficult for some of
his copyists to begin a new notice under 1148. 1231 instead of 1225
for the Egyptians may be the 9th year of Psammis instead of the 9th
year of Osorthon : such a change could be made the more easily if the
notice originally stood in the margin, against the Egyptian column. It
is interesting to observe that Scheme quotes ABFP in favour of this passage
standing in the margin in Jerome's version instead of in the column for
events. The substitution of 1206 for 1286 admits of an equally simple
explanation. Jerome has erroneously entered the death of Bocchoris at the
hands of Sabacon against the first year of Sebichos (1291) instead of against
the first year of Sabacon (1282). The Carian thalassocracy, which is the next
entry, appears to have been shifted along with the death of Bocchoris. Thus
it retains its proper place in relation to other entries in the same column, but
is dated ten years too late. That both errors were made by Jerome, not
Eusebius, is proved by a reference to the Armenian version, where both
entries appear in' their proper place.

So far then as each successive thalassocracy is noted in Jerome's Canons,
there is no difficulty in restoring the text of Eusebius and explaining
the errors that have crept into it. The remainder of the series can best
be restored by beginning at the end and working backwards. It is clear
from the words in which the last thalassocracy (that of the Aeginetans)
is entered in the Canons, closing with ' usque ad transitum Xerxis,' that the
list was meant to end with the expedition of Xerxes. This last thalassocracy
lasted ten years, as is proved by the concurrent testimony of the Armenian
Chronographia, the Armenian Canons, and Syncellus, and it therefore
follows that Jerome's version is seriously in error in making it begin in
1508 (508 B.C.). The date given in the Armenian Canons, 1531 ( = 185 B.C.).

:actly ten years before the date of Plataea and Mycale as preserved
in Jerome, though only five years before the date assigned to the destruction
of Athens. It is therefore consistent with the principles on which Eusebius

7 'Tumultuarium opus.' P. 1. 1. 14 (Scl.

ON THE 'LIST OF THALASSOCRACIES' IN EUSEBIUS.

arranged his chronology, and, as we shall - se presently, it is confirmed by
other figures. In Syncellus the notice of this thalassocracy immediai
follows a notice of Gelon, which is placed in Jerom<' againsl the year 1530,
a piece of evidence of little importance in itself, but valuable as confirming
the Armenian <'<m>iis. Taking this date as a starting point, ami working
with the figures of the Armenian Chronographia, which we have found to be
confirmed by figures derived from the Canons, we obtain the following
series : —

Duration

"f Abraham

X 1 V.

Lacedaemonii.

■2

1504

512

XV.

Xaxii.

1506

510

XVI.

Eretrienses.

1.".

1516

500

XVII.

Aegim

l".:;!

•'

This gives us tor the Lacedaemonians a date differing by one year
•only, and for the Eretrians a date differing by two years only, from those
of the Armenian Canons. As we have already seen, small errors like these
are the almost necessary result of the method in which tin s are

constructed. It is also important to observe that the date thus obtained
for the Naxians differs by two years only from Jerome's date for the Aeginetan<.
This renders it probable that Jerome or his amanuensis, after taking his
eye off his Greek original, while red ink was being substituted for black.
allowed it to be caught by an entry similar to the one which he wished
to insert, and so the Aeginetan thalassocracy was entered where the Naxian
thalassocracy should have been. It is also possible that the entry may
have been made either in Jerome or in the Greek manuscript used by him
under 1508 instead of 1506, because a long entry under 1505 occupied
the whole space belonging to the years 1505, 1506, and 1507.

It is difficult to determine with certainty the dates assigned by Eusebius
to the Phocaean and Samian thalassocracies. The dates 1441 and 1486 in
the Armenian Canons agree so well with the 41 years' duration of the
Phocaean thalassocracy that it is impossible to suppose that they contain any
serious error. The date 1441 receives confirmation as an approximate date
from Syncellus, who mentions the Phocaean thalassocracy immediately after
the seven wise men (1438 in Jerome) and immediately before the Nemean
games (1444 in Jerome). We are therefore compelled to abandon Jerome's
€8 years for the Lesbians, while the '.Hi years of a single Armenian manuscript
are, as Mr. Myres observes, calculated from the dates assigned bo the
thalassocracies in the Armenian version, and arc therefore ofn i value for the
text of Eusebius. At the same time we have no evidence for the duration
•of the Samian thalassocracy apart from the date assigned to it in the
Armenian Canons. Accordingly, the dates 1441 and I486 cannot be checked
by means either of earlier or of later dates in the series, an 1 as the Armenian
Canons cannot be trusted for an exact year, both dates may contain
a slight error. In any case, one of the two must be wrong, since, a- they
stand, they would give a duration of 45 instead of 44 yens for the Phocaean

86 J. K. FOTHERINGHAM

thalassocracies. The Samian thalassocracy is certainly connected with the
tyranny of Polycrates, which Jerome dates 14S4, S but, as Mr. Myres observes,
this date was probably taken by Eusebius from some source other than his
list of thalassocracies. It is also interesting to observe that the notice of the
Samian thalassocracy appears in the Armeuian Canons between that of
Pythagoras (1487 in Jerome) and that of Hipparchus (1489 in Jerome). By
combining these last two dates with those quoted above from Syncellus for
the Phocaean thalassocracy, we get 1443 and 1487 or 1444 and 1488 as the
dates of the Phocaean and Samian thalassocracies, of which the former pair
most closely approximate to the dates given in the Armenian Canons ; but
the calculation is rather precarious, and we must be content to leave
the exact dates assigned by Eusebius to these thalassocracies an open
question.

We are now in a position to complete our table of thalassocracies.
leaving the exact duration doubtful in two instances. The complete list will
be as follows : —

Duration

uaie in years
of Abraham

Date B.C.

Lydi.

844

1172

II.

Pelasgi.

936

1080

III.

Thraees.

1021

995

IV.

Rhodii.

1100

916

Phiyges.

1123

893

VI.

Cyprii.

1148

868

VII.

Phoenicia

1180

836

VIII.

Aegyptii.

1225

791

IX.

Milesii.

1268

748

Cares.

1286

730

Lesbii.

96 (?)

1347

669

XII.

Phocaeenses.

1443 (?)

573 (?)

XIII.

Samii.

17(?)

1487 (?)

529 (?)

XIV.

Lacedaemonii.

1504

512

XV.

Xaxii.

1506

510

XVI.

Eretrienses.

1516

500

XVII.

Aeginetae.

1531
-1541

485
-475

697

It is i in) mil ant to notice that the series thus obtained is calculated
entirely from figures derived ultimately from the Canons, the Chronogrdpkia
having been used for corroboration only ; but as the series accords almost
exactly with the extant figures of the Chronographia, it is probable that
it correctly supplies the figures missing from the surviving manuscripts.
If we wisli to go behind FAisebius to his authorities, we must set aside
tin- dates given above and concentrate our attention on the durations, and
we must remember thai the scries ought to end, not in 475 B.C., but in
480 B.C., the true date of the expedition of Xerxes. It ought in consequence
to extend back to 1177 j:.c, only four years after the date assigned by

So the Bodleian and Berlin manuscripts |M<> .

ON THE -LIST OF THALASSOCRACIES ' IN EUSEBIUS. 87

Eusebius to the fall of Troy; and one is tempted to suppose that the
compiler gave some explanation of the three intervening year3, so as to
bring Ins List back to the fall of Troy itself, and to complete a total of
700 years.

It will be observed that from the Phocaeans onwards the duration of
each thalassocracy, according to the figures given in the table above,
corresponds with its duration as obtained by Mr. Myres. Nor is it possible.
to dispute the historical facts with which Mr. Myres connects each of these
thalassocracies. It will also be observed that there is a tendency for a thalasso-
cracy to end in some disaster to the power that held it. Thus the Phocaean
thalassocracy ends in the capture of Phocaea by the Persians in 534 B.C.,
the Samian thalassocracy with the death of Polycrates in or about 517 B.C.,
and the Eretrian thalassocracy with the fall of Eretria in 490 B.C. The
Lesbian period should cover the years 674-578 B.C., the period in which the
ancients were accustomed to place the glorious names belonging to Lesbos,
including Lesches at the beginning, followed by Terpan ler, Arion, Pittacus,
Sappho, and Alcaeus. It is not easy to explain a Carian sea-power in
735-(>74 B.C., but it may be connected with the Carian mercenaries in the
service of Psammetichus. There is,' however, no difficulty in explaining a
Milesian thalassocracy in 753-735 B.C. It is the age of colonization. The
beginning of the thalassocracy is connected in Eusebius with the foundation
of Naucratis, and is immediately followed by a notice of Thales. The notice
of the foundation of Trapezus 9 in 755 B.C. probably belongs to the same
system of chronology, with the accuracy of which we are not concerned. It
is important to note that the foundation of Naucratis in Egypt is the starting-
point of this thalassocracy. The compiler seems to have regarded this event
as a symptom of the downfall of Egyptian sea-power and the establishment of
a Milesian power in its place, and it is significant to observe that the
Milesians are in the list preceded by the Egyptians. We have thus three
successive thalassocracies, the Egyptian, the Milesian, and the Carian,
assumed from evidence (more or less slight) of power in the Egyptian
delta.

The importance of the Egyptian thalassocracy seems to lie mainly in its
downfall. The period assigned to it, 796-75:} B.C., is not marked by any
events in Egyptian history, famous among the Creeks, except perhaps the
reign of the Egyptian Hercules (799-792 B.C. in Eusebius) and the reign of
Bocchoris, who seems to have enjoyed a celebrity .piite out of relation to his
real importance, and whom Eusebius dates 779-735 B.C. It is interesting to
observe that the 4:> years which Eusebius assigned to this thalassocracy,
according to the text as restored in this article, correspond closely with the
44- years which he attributes to Bocchoris. On the other hand, the date-
both of the Egyptian Hercules and of Bocchoris profess to be based on
Manetho, whose system of chronology dors not seem to have come into
general use till after the time of Diodorus. from whom the list o\' thalas-

the Armenian version and Synccllus. Jerome has '< ■

88 J. K. FOTHERINGHAM

socracies is derived. It may therefore be necessary to set aside these
synchronisms as mere coincidences.

With the Phoenician thalassocracy we appear to be once more upon safe
ground. The period 841-79G B.C. contains the dates assigned to the
foundation of Carthage by Trogus Pompeius, Velleius Paterculus, Timaeus,
ami Servius. 10 It ought also to include the date cited by Josephus 11 from
Tyrian sources. He gives a detailed scheme of Phoenician chronology from
Hiram to Pygmalion, and a total of 143 years and 8 months for the period
from the building of the temple at Jerusalem to the foundation of Carthage.
According to the received chronology, Solomon's temple was begun in 1012
and completed in 1005 B.C., but it is clear from Assyrian references to Ahab
and Jehu that these dates are 43 years too high. This would reduce the
Tyrian date of the foundation of Carthage to 1012 or 1005-43-143 = 826 or
819 B.C. The period in which the most famous event of Phoenician
maritime history falls was naturally made a period of Phoenician thalas-
socracy. But before the Phoenician period all is darkness. The only
tangible event that it seems possible to connect with any of the earlier
sea-powers is the Thracian conquest of Bebrycia or Bithynia, which
Eusebius places in 972 B.C., and which therefore falls within the Thracian
thalassocracy. It is at all events clear that the list as Eusebius found it was
correlated to the general chronological tradition which he follows, and the
presumption is that the list was arranged to conform to some of the better
known dates in this tradition and that other dates were made to conform to
the list.

It is difficult to say when the list was drawn up. Mr. Myres's arguments
for a fifth century date are far from convincing ; our list bears little resemblance
to any scheme presupposed in the narrative of Thucydides. 1 ' 2 His sketch
of nautical history includes western as well as eastern sea-power, and where it
gives dates they do not tally with those of our list. He begins with a
Corinthian sea-power, and mentions the building of triremes for the Samians
by a Corinthian ship-builder at a date (704 B.C.) which the list places in the
Carian period. The Corinthian sea-power would seem, according to his
narrative, to have been continuous. It is true that he knows of Ionian sea-
power ' later ' than the Corinthian in the time of Cyrus and Cambyses, thus
tacitly denying a Milesian sea-power in the eighth century B.C. Of the two
Ionian states that he mentions, the Samians are indeed connected both by
Thucydides and by the extant list with the reign of Cambyses, but the only
event that he connects with the Phocaeans by name, the founding of Massalia,
falls outside the period of their sea-power as given in the list. It is possible
that the survey of sea-power in Thucydides may have suggested the idea of
framing a list of thalassocracies. In that case the present list, which is not
made to conform to Thucydides, is not the earliest. But it seems vain to
pursue the inquiry further. This article will have served its purpose if its

10 See the table in Smith, Dictionary of » <•. Ap. i. 17, 18.

■Greek ami Hi,, mm Geography, i. 531. <- j. 13.

ON THE 'LIST OF THALASSOCRACIES ' JN EUSEBIUS. 89

attempt to determine the figures in the list which Eusebius derived from
Diodorus is successful. I do not regard the list as of any great historical
value. There may be something behind it which has not yet been discovered ;
but inquiries into its value are bound to be fruitless, unless the restoration of
the list is regarded as a problem of textual criticism, independent of historical
explanations.

J. K. FOTHERINGHAM.

PEPARETHUS AND ITS COINAGE.
[Plate IV.]

The Plate which illustrates this article represents a series of silver
coins, Nos. 1-8, which bear a strong family likeness. They are all tetra-
drachms of the Euboic standard, and each displays on the obverse a peculiar
bunch of grapes which would have excited even the infantine ridicule of the
humblest painter of Dutch fruit-pieces. Two of these coins, PI. IV. 2 and 8,
both recent acquisitions of the British Museum, are unpublished, and present
new types. Nos. 9-1 1 are bronze coins of Peparethus.

In type and style the most remarkable of these coins is PI. IV. 1 =
Fig. A. This piece was found in the island of Cos, but Mr. Bar-clay Head,
who first made it known in 1891, 1 attributed it in a very ingenious and learned
paper to Cyrene. This attribution carried with it the assignment to Cyrene of
the types here figured as PI. IV. 8 (and 5), 4 (and 6), because all three
types were found to be united together by a concatenation of dies. In a
paper printed in the Numismatic Chronicle, 2 shortly after Mr. Head's, I ven-
tured to point out — without suggesting any better attribution — that the great
difficulty in this classification was that all the known coins of the African
city bore as their type the silphium-plant, or, at least, a leaf or seed
representative of that plant : it seemed difficult to fit into the Cyrenian
currency a bunch of grapes, a winged figure, a helmet and a head of
Herakles. A distinguished numismatist, M. Waddington, to whom at the
time, I mentioned the proposed attribution, told me that he thought, in
spite of Mr. Head's excellent article, these grape-coins would turn out to be
Euboean or Macedonian.

Writing in this Journal in 1897 Mr. Hill suggested the Macedonian
( li.ilcidice as the probable home of the coins. About 1904, a specimen with
the helmet reverse was procured by its owner near Salonica. Subsequently
Mr. A. J. B. Wace obtained in Scopelos, i.e. the ancient Peparethus, the
island lying beyond the coast of Thessalian Magnesia, specimens of the
helmet and Herakles reverses; and in 1900 another specimen of the
Herakles was shown at the British Museum by its owner, who stated that
it was found in the same island.

1 Num. Chron. 1801, p. 1. - 1892, p. 20.

J. H. S. VOL. XXVII (1907) PL. IV.

...

liSfc

1, /r

COINS OF PEPARETHUS, &c.

PEPARETHUS AND ITS COINAGE. 91

Finally, in L906, the British Museum acquired the coin PI. IV. 2 =
Fig. C. This coin bears the letters PE already seen on the bronze coins of
Peparethus, displays like them Dionysiac types, and was found in Scopelos.
Irs attribution to Peparethus is, thus, hardly open to doubt, and the coin has
numismatic importance as showing that this island coined silver, as well as
bronze money,and that its coinage began somewhat early in the fifth century.

Unfortunately this inscribed coin cannoi be held to pn ve that the grape-
coins previously referred to belong to Peparethus, for the bunch of grapes on
its obverse is not identical with the bunches on the other coins. It seems,.
however, to strengthen their attribution to this island, an attribution first
suggested by the provenance of some of the specimens. In this paper I shall
therefore venture to adopt as a probable hypothesis the Peparethian origin of
all the grape-coins, except, perhaps, in the case of the coin PL IV. 8.

The island which chose the bunch of grapes as its principal badge,
though less famous than Naxos or Th era, was in legend declared to have been
colonized by Cretans under an appropriately named leader, Staphylos, the
son of Dionysos and Ariadne. Dionysos was its principal divinity, and
Staphylos is still the name of a bay of the island. 4 It is first mentioned in
the Homeric Hymn to Apollo. In antiquity it was well wooded; it grew
corn and olives and exported a well-known wine. 5 At the present day it
sends a light, red wine of its production to Constantinople and the Black
Sea ports. There were three towns in the island, namely, Peparethus, the
most important place, now called, like the whole island, Scopelos; Selinus,
and a harbour-town, Panormus.

Coming now to the more precise attribution of our coins, it is probably
not rash to assume that the chief minting-place was the town Peparethus.
We may assign to it the inscribed coin with the seated Dionysos, No. V. 6
(PI. IV. 2) and the bronze coins, PJ. IV. 9—11. One would suppose also that
the coin, PI. IV. 1, with a fine figure-subject (Fig. A) was likewise issued from
the same mint. But to what mint are we to assign the Herakles-head, the
helmet and the ivy-wreath, each of which has a bunch of grapes as its
obverse ? There seems some difficulty in assigning so many reverse-types to
the same town during a period (apparently) of about forty or fifty years; aud

Peparethus is no1 the only provenance 4 On Peparethus, see Bursian, Geographic

recorded for these ruins. They have been oon Grieehenland, ii. pp. 386 f. : C. Fredrich,

found in Cos, in Macedonia (Salonica), and in 'Skiathos und Peparethos' in ifiltheil. arch.

Thessaly. Cos may safely be ruled oul ns the Inst. (Athens) x\\i. 1906) p. 99 f. and refer-

mint-place of these coins, for they in noway enees there; cp. "Wace, lb. p. 129 f. •Skiathos

amalgamate with the already well-known series und Skopelos ' ; Murray's '

of Coan money. Some coast-town ofThessaly, pp. 931 f.

or, better, of Macedonia would have a fair 3 On the wine. Demosth. In La

claim to the coins, if the attribution to Soph. Philoct. 548 ; Heracl. Pont Fragm. 13 ;

Peparethus is unacceptable. The helmet- Athen. i. p. - J0 a and f; Pliny, //. .V. xiv.

reverse is rather distinctively Mar, ■.Ionian 7. 76.

and the winged figure (Fig. A) lias been ,; The Roman numerals refer t<> the des^rip-

compared (by Mr. Hill, J.ff.S. 1897, p. 7'." tive list of the coins given at -the end of this

with the winged figure with a wreath on a ait:
Macedonian | | coin.

92 WARWICK WROTH

I at first thought that the Herakles and helmet types — specimens of which
were ascertained by Mr. Wace to have been found on the site of Selinus —
might be attributed to that town. This provenance, however, in a small
island, is not decisive as to origin, and bearing in mind the way in which
this series of coins is linked together by the interchange of dies, I think the
safest course is to suppose that they were all struck at a single mint-place,
namely the town Peparethus.

The only exception may be the coin No. VI. (PI. IV. 8). It has a
specially marine character. On the obverse, four dolphins are added to the
simple bunch of grapes, and the reverse is a dolphin-rider. These types
might suit the harbour-town Panormus. but, on the other hand, this coin — at
present unique — is stated to have been found in Thessaly, at Demetrias, near
Volo, and the addition of the dolphins to the bunch of grapes, which seems
to be the badge of Peparethus, rather suggests that the coin does not belong
to this island but to an adjacent island or, perhaps, to some coast-town of
Magnesia, where Dionysiac types are already known from the coins.

Date and types. — I have already mentioned that several coins in our
series show a curious concatenation of dies. The importance of systematically
studying the identities of dies has lately become more "widely recognized :
Dr. Regling, for instance, in his recent admirable monograph on the coinage
of Terina has carefully noted the relationship of the various dies, and gained
thereby good clues to the exact chronological sequence of the coins. In the
present case, it will be found that the die for the grapes-obverse of the
winged figure coin (No. I.) has been used for the obverse of a Herakles coin
(No. II. A). Again, the grapes-die found in conjunction with the Herakles
coin, II. D is used as the obverse of III. (the helmet type), and also for the
obverse of IV. (the ivy-wreath type). The relation of the dies may be set
forth as follows, identical letters indicating identical dies: —

No. I. (Winged figure) Obv. a, Rev. a.

II. A. (Herakles) ,, a, „ B.

„ II. D. (Herakles) „ j3, „ b.

„ III. (Helmet) „ /?, „ C.

.. IV. Ivy- wreath) „ /3, „ D.

Tliis examination of the dies suggests that Nos. I. and II. A are nearly of
the same date. In the Herakles-series, II. D is later (but not much later)
than II. A. The helmet-type III. and the ivy-wreath type IV. are both
contemporary, or nearly contemporary, with the Herakles-type II. D.

No. V. seated Dionysosj and No. VI. (dolphin-rider) do not share in
this interchange of dies.

No. I. (PI. IV. 1 = Fig. A). Mr. Hill has well suggested 7 the name of
Agon — a male personification corresponding to Nike — for the reverse type.
This little running figure is executed with all the minuteness of gem-

■ J.H.S. 1897, p 80.

PEPARETHUS AND ITS COINAGE. 93

engraving, but is full of elastic vigour. In some respects it recalls the
Poseidon of the coins of Poseidonia 8 and has some affinities with the Nike of
of Elis 9 and the running Nike of Cyzicus. 10 All these are early coins, and I
think our Agon can hardly be later than circ. B.C. 500-490. The type is,
apparently, not Dionysiac, and it is hard to suggest the reason of its choice.

No. II. (PI. IV. o and 5 ; Fig. B). The Herakles head has an aspect
unusual on coins. It is delicately treated but has the bulging eyes and
simpering smile found in representations of Herakles on early vases and other
monuments. It may be compared in style with the Dionysos head on an
archaic coin of Sicilian Naxos. 11 This type (in its earliest manifestation,
PI. IV. 3) must be placed soon after the Agon coin (No. I.), of which it has
borrowed the obverse-die. Perhaps the date is circ. B.C. 490. The variety of
this coin (PI. IV. 5), where the obverse-die is changed so as to present three
bunches of grapes, must be somewhat later, B.C. 490-485 (?).

No. III. (PI. IV. 4 and 6) introduces a new reverse — the helmet, but
the obverse die is borrowed from No. II. D (Herakles). We may date it,
approximately, B.C. 485-480. The significance of the helmet is not obvious:
it is a type that is chiefly familiar on Macedonian coins. V1

No. IV. (PI. IV. 7). The reverse does not, so far as I know, find a
parallel in any other coin-type. We might be content to explain it as a
mere Dionysiac emblem, but it may perhaps be preferably described as a
votive wreath. We know from Athenaeus 13 that the Peparethians dedicated
at Delphi a golden ivy-wreath — arefyavov y^pverovv kittov UeTrapyOicov. The
date of this dedication is not known: in the same sentence some other
Delphian dedications are recorded including a laurel-wreath of the Ephesians
and four golden aTXeyytSia offered by the people of Sybaris. If we could
assume that all these anathemata were made on the same occasion, the date
of the offering could be approximately fixed as not later than B.C. 510, the
date of the destruction of Sybaris. This ivy-leaf reverse is joined with an
obverse-die borrowed from the helmet-coin (No. III.) so that it must be
nearly contemporary ; circ. B.C. 480 (?).

No. V. (PI. IV. 2 ; Fig. C). Seated figures are rare on archaic coins and
even until the acre of Alexander the Great : notable instances are the seated
Zeus of Aetna, 14 circ. B.C. 476-4(31 and the Harmonia (?) seated on a diphros
on a coin of Thebes, circ. B.C. 446. 15 Our seated Dionysos shows the heavy
treatment of the figure found on archaic sculptured reliefs of a similar

v Sead, ■ to Coin i of Ancients, PI. VII. II. ..

12. '- Svoronos, Journ. internal. <!'<rrch. Num.

■ V. Gardner, Types, PI. III. 14. 190."., p. 341.

"' Wroth, B. M. Cal Mysia, PI. IV. 7 '• Theopompua ap. Atlien. xiii. 605 is. ...

and 9. quoted 1 • \- Rouse, Greet e w, p. 281.

11 Hill C< j ily, PI. I. :'.. Cp. li Hill, Coins of . 6 ily, PI. IV. 13.

also the head <>t' a warrior on a Lycian coin. '"' Head, Guide to Coins of Ancient*, 1 I.

B.C. 500 160 in Hill, P. M. Cat. LyAa, PL XIII. 15 and B. M. p. 72.

94 WARWICK WROTH

character, and might, independently of its obverse, be placed very early in
the fifth century. It is accompanied however, by an obverse which is plainly
later than any of those previously described, for on this obverse the bunch of
grapes is less crudely represented and an inscription (PE) makes its
appearance. I Avould therefore date the coin circ. B.C. 480-470.

The coin No. VI. (PI. IV. 8) as I have already remarked is probably
not of Peparethus. The four dolphins encircling the bunch of grapes were
presumably suggested by the coins of Sicily, on which they appear first, at
Syracuse, in the time of Gelon, i.e. circ. B.C. 485. 1(i The dolphin-rider on
the reverse is not satisfactorily preserved, but I am inclined to think that
it is a female figure wearing a long chiton like Europa on her bull on the
metope of Selinus or on the early coins of Cnossus : a male dolphin-rider is
already known from an early coin (sixth or seventh century ?) attributed by
Svoronos to the island of Syros. 17 The date of our coin may be provisionally
fixed as circ. B.C. 480.

Between circ. B.C. 470 and 400 there is a broad gap in the coinage of
Peparethus. During this period the island was no doubt subordinate to
Athens. 18 The bronze coin IX figured PL IV. 9 may be placed circ. B.C. 400.
It displays a bearded head of Dionysos of good style. No. X. (PI. IV. 10)
shows a beardless Dionysos, perhaps of the third century.

No native coinage can be assigned to the island during the fourth
century. In B.C. 377 the Peparethians are named among the allies of the
Athenian Confederacy. In B.C. 361 the town of Peparethus was besieged by
Alexander tyrant of Pherae and it is interesting to know that two of his silver
coins have been discovered in Scopelos. 19 The island was afterwards laid waste
by command of Philip II. of Macedon because the Peparethians had seized
the island of Halonesus. At the end of the third century (B.C. 209-200) the
town was contended for by Philip V. of Macedon, by Attains of Pergamum,
and by the Romans. The coin (PL IV. 11) doubtless belongs to the second
or first century B.C. The worship of Athena, whose head appears on it, is
known from other sources to have prevailed at the towns of Peparethus and
Selinus.

In conclusion, I set forth the details that will be looked for by
numismatic readers, some of whom may be able to carry farther than I have
done the dating and attribution of this interesting but rather difficult series
of coins.

1,1 A single dolphin was the badge of Ceos British Museum in March 1906. A note of

ami mi tin- coins it accompanies the distinctive these was made by Mr. Hill, as follows: —

types of the various towns of the island. 1. OLc, Head of Hecate r. hair rolled ; in front,

17 Journ. internal, d'areh. num. 1900, p. 59 ; arm holding torch. Rev. AAEEANA Lion's

cp. Head, B. M. Cat. Carta, p. lix. head r. ; below, double-axe, JR Size - 7. Wt.

11 Sir /. <7. I. Index of Athenian Tributary 87 '2 grains (similar to B. M. Cat. Thessaly,

Allies. PI. X. 12). 2. Obv. Wheel. Rev. A AE double-

19 Two coins {dans le commerce) shown at the axe. At Size '4. Wt. 12 6 grains.

PEPARETHL'S AND ITS COINAGE.

I. — Winged Figure Type.

Obv. Bunch of grapes. Border of dots.

Rev. Winged male figure (Agon ?), naked, running r. ; wears boots with

tags ; in each hand, wreath. Square compartment of dots. Whole in

incuse square.

M Size 1-05. Wt. 261 grains. PL IV. 1 and Pig. A. In British
Museum, acquired in 1891 (B. V. Head, Num. Chron. 1891, p. 1 ; PI. I.

FlG. A. (Scale 2 : 1.)

(Scale 2:1.)

3 ' Cyrene '; Wroth, Num. Chron. 1892, p. I!) ; Hill, J.H.S. 1807, p. 79,
'Chalcidice' ?). Found in Cos together with Nos. II. B and III. B, an
archaic tetradrachm of Athens, and an archaic tetradrachm of Mende.
Obverse from same die as No. II. A and II. B and II. C.

II. — HeraJdcs Type.

Obv. Bunch of grapes. Border of dots.

Rev. Head of bearded Herakles 1. in lion's skin. Square compartment of
dots. Whole in incuse square.

A. In Sir H. Weber's Coll., London. JR, Size -95. Wt. 265-5 grains.
PL IV. 3 and Fig. B. Procured from Greece. Obv. from same die as
No. I.

B. In Museum of Fine Arts, Boston (originally in the Greenwell Coll.).
M Size •!). ' Wt. 2643 grains.' (B. V. Head, Num. Chron. L891, p. 1,
No. ii. PI. I. 4 = Regling, Sammlung Warren, No. 1410; PI. 32,
Fig. 1410. The wt. is there stated to be 16'76 grammes.) Found in
Cos with No. I.

Obv. from same die as No. I. (Obc. and rev. same dies as II A

96 WARWICK WROTH

C. la A Private Collection. M Size 95. Wt. 17T 5 grammes. (Svoronos,
Journal internal, d'arch. num. 1905, p. 339, No. 4; PJ. XL 22.) Found
on flic site of Selinus in Peparethus and obtained in the island by Mr.
A. J. B. Wace ; the statement, Svoronos, I.e. p. 340, that this coin (and
III. A. infra) were found in Skiathos, is based on a misapprehension.
(Information from Mr. Wace.)

Obv. from same die as No. I. Obv. (and rev. ?) same dies as II. A
and II. B. •

The obv. of II. D and E next to be described differs from the obv r
of II. A, B, C, in having a small bunch of grapes on each side of the
large bunch.

D. In British Museum, acquired in 1872 from Edward Wigan's collection.
M Size 1. Wt, 256 grains. PL IV. 5. (Head, Num. Chron, 1891, p. 1,
No. iii. PI. I. 5.) Provenance unknown.

Rev. from same die as II. A and II. B.

E. Dans le commerce, 1906. JR Euboic tetradrachm. Found in Scopelos
(Peparethus).

Obv. and rev. from same dies as II. D.

III.— Helmet Type.

Obv. Bunch of grapes, flanked by two smaller bunches. Border of dots.
Rev. Crested Corinthian helmet r. within incuse square.

A. In A Private Collection. M tetradrachm. Wt. 16*75 grammes.
(Svoronos, I.e. p. 339, No. 1, PI. XL 19.) PL IV. 4. Found on the site
of Selinus in Peparethus, and obtained in the island by Mr. A. J. B.
Wace. (Information from Mr. Wace.)

Obv. from same die as No. II. D.

B. In British Museum, acquired 1891. M Size 1-05. Wt. 253-4.
(Head, Num. Chron, 1891, p. 2, No. iv. PI. I. 6; wt. stated as
261-3 grains.) PL IV. 6. Found in Cos with No. I. etc.

Obv. aud rev. from same dies as No. III. A.

C. In A Private Collection ? M tetradrachm. Wt. 16 - 50 grammes.
Found, by a native of Thessaly, 77700? votov tov 'OXv/jlttov. (Svoronos, / c.
p. 339, No. 2 ; PI. xi. 20.)

Obv. and rev. from same dies as No. III. A.

D. Dans le commerce. JR tetradrachm, shown at the British Museum in
1904. Obtained near Salonica.

I V. — Ivy-wreath Type.

Obv. Bunch of grapes Hanked by two smaller bunches. Border of dots.
(Flaw in die, on r.)

PEPARETHUS AND ITS COINAGE.

Rev. Ornamental device consisting of a pellet surrounded by dots and four
ivy-leaves arranged diagonally. (Votive ivy-wreath ?) Square compart-
ment of dots. Whole in incuse square.

In A Private Collection. M tetradrachm. Wt. 1768 grammes.
Found by a native of Thessaly, 7rpo? votov tov 'OXv/ulttou. (Svoronos, /.c.
p. 339, No. 3 ; PL XI. 21.) PI. IV. 7. (Casts of this and of No. III. A
have been kindly supplied by M. Svoronos.)

Obv. from the same dies as Nos. II. D, E.

V. — Seated Dionysos Type.

Obv. Bunch of grapes, with slight indications of smaller bunch at each
side. On 1., PE (the E repeated through double-striking). Border of
dots.

Rev. Dionysos with long beard and long hair, seated 1. on diphros ; himation
over lower limbs; in outstretched r., kantharos; 1. hand, resting on side,

Fig. C. (Scale 2:1.)

holds thyrsos. Square compartment of dots. Whole in incuse square.
M plated with silver. Size P05. Wt. 220*3 -rains. PI. IV. 2 and
Fig. C. In British Museum, purchased in 1906.

Found in Scopelos (Peparethus).

VI. — Ralph in -rider Type.

Obv. Bunch of grapes, around which four dolphins swimming.

Rev. Figure riding 1. on dolphin (apparently a female figure wearing long

chiton girt at waist); the type within an incuse square to which it is

adjusted diagonally.

In British Museum, purchased (together with No. V.) 1906.
M Size 1*2. Wt. 259 grains. PI. IV. 8. Found at Demetrias in
Thessaly.
H.S. VOL. XXVII. H

98 PEPARETHUS AND ITS COINAGE.

VLI.-IX. — Bronze Coins of Peparethus.

VII. Obv. Head of bearded Dionysos r., wreathed with ivy.

Rev. P E Kantharos; wreathed with vine-leaf and two bunches of
grapes; circular incuse.

British Museum, acquired in 1906. M Size '55. Found'
in Scopelos. PL IV. 9.

VIII. Obv. Head of beardless Dionysos r., wreathed with ivy.
Rev. P E Kantharos.

British Museum, acquired in 1891. M Size ^o..
Obtained in Greece? PI. IV. lO.

IX. Obv. Head of Athena r. in helmet.
Rev. n E Bunch of grapes.

n a

British Museum, acquired in 1891. M Size - 55.
Obtained in Greece? PI. IV. 11.

(For other bronze coins of Peparethus, see Gardner, B.M.
Cat. Thcssaly, etc. s. v. Peparethus; Macdonald, Hunter Cat. I.
p. 460.)

Warwick Wroth.

J. H. S. VOL. XXVII. (1907). PL. V

FIGURE B FROM THE COOK SARCOPHAGUS

J. H. S. VOL. XXVII. (1907). PL. VI

FIGURE C FROM THE COOK SARCOPHAGUS.

J. H. S. VOL. XXVII. (1907). PL. VII.

FIGURE D FROM THE COOK SARCOPHAGUS.

J. H. S. VOL. XXVII (1907). PL VIII

FIGURE E FROM THE COOK SARCOPHAGUS.

J. H. S. VOL. XXVII. (1907). PL. IX.

FIGURE F FROM THE COOK SARCOPHAGUS,

J. H. ?. VOL. XXVII (1907). PL. X.

FIGURE G FROM THE COOK SARCOPHAGUS.

J. H. S. VOL. XXVII. (1907). PL. XI.

FIGURE H FROM THE COOK SARCOPHAGUS.

J. H. S. VOL. XXVII. (1907). PL. XII.

FIGURE J FROM THE COOK SARCOPHAGUS.

A SARCOPHAGUS OF THE SIDAMARA TYPE IN THE

COLLECTION OF SIR FREDERICK COOK, BART., AND

THE INFLUENCE OF STAGE ARCHITECTURE

UPON THE ART OF ANTIOCH.

[Plates V.— XII.]

In the collection of Sir Frederick Cook at Doughty House, Richmond,
are nine fragments of a large sarcophagus of surpassing interest for the
modern scientific history of art. They were identified some two years ago by
Mrs. Arthur Strong, who is engaged upon a new illustrated catalogue of the
Richmond antiques 1 which is shortly to appear in this Journal. Mrs. Strong
at once communicated to me the existence of a sarcophagus of so much
importance for the studies I had initiated in my book Orient oder Rom, and,
by the courtesy of the owner, I at the same time received a set of
photographs of the nine fragments.

I shall first describe the fragments, determining at the same time, by
comparison with kindred examples, the Art group to which they belong, and
shall then endeavour to make clear their significance.

Of the nine fragments, which are all about the same height (3 ft. 3 in.)
and thickness, eight are decorated with single figures executed in high relief
and almost detached from the background, resembling, in fact, statues in
niches. One fragment alone, which I will take first, forms an exception.

A. — Fig. 1 (height 2 ft. 7h in., breadth 1 ft.). Here we see a door, in the
opening of which is a table standing on four lion's feet; upon this is a
circular altar, whence flames seem to rise. Over the door-lintel with its
straight moulding is a |"~~| shaped upper member that displays, under
broad dentils, a scroll-work of peculiar character. This panel supports a
projecting impost and is decorated with the same scroll-work disposed

1 The Richmond collection is, of course, in- ments. According to Sir Frederick Cook's

eluded in Michaelis' Ancient }farblcs in Great house-steward they had been for over fifty

Britain. At the time when this book was years in a garden in London, and were quite

published, the sarcophagus now described was black with London dirt when they were first

not yet in the collection. It was purchased by brought to Richmond. Till two years ago two

the' late owner, Sir Francis Cook, about twenty- of the fragments remained in the Doughty

two years ago. Mrs. Strong has been unable House Conservatory.
to discover the previous history of these frag-

II 'I

100 JOSEF STRZYGOWSKI

symmetrically to the centre. The door thus described stands in front of a
pediment, which rests on spiral-fluted columns. Below the pediment is a
scallop-shell. Above this shell appears an architrave broken in the centre.
Like the geison, it is composed of a fillet from which hang irregular dentils.

Fig 1.— Fragment A of the Cook Sarcophagus.

It is filled up by two huge members borrowed from an egg-moulding and
these reappear on each side of the shell. The pediment terminates at the
top with a border upon which a tendril is carved, not in relief, but in a
new coloristic manner. Above the fillet are horizontal oo-shaped Acroteria.

A SARCOPHAGUS OF THE SIDAMARA TYPE

101

There exists a whole series of sarcophagi, all of which display on one
narrow end the same door in front of a shell pediment, and also correspond
in ornament and in technical execution with the fragments at Richmond.
We can best form an idea of what the Richmond sarcophagus, when

Fig. 2.— The Sidamara Saro iphagus.

complete, was like by reference to the huge sarcophagus d at

Ambar-arassy, 125 kilometres west-south-west ofKonia the ancient [conium)

in Asia Minor, now in the Imperial Ottoman Museum

and known as 'the Sidamara Sarcophagus' {Monuim U V ', ix. Plates

XVI1.-XIX.).

102 JOSEF STRZYGOWSKI

Fig. 2 shews all the details described in Sir Frederick Cook's fragment.
Here, however, the central design is flanked on the left by a woman, on the
right by a man, and the whole is enclosed within two columns surmounted by
tall imposts. Similar examples are to be found also in the Louvre, 2 in the
Giardino Colonna in Rome, 3 at Athens, 4 and at Ueskeles in Asia Minor. 5

B. — PI. V. (breadth 2 ft. 1 in.) reproduces the first of three nude figures
of youths. The figure is shewn standing in front view; the weight is on the
right leg, and the left leg is at ease ; the head is turned in profile to the
right. Long curls fall down to the shoulders, over which is thrown a
chlamys. The left hand catches the drapery up in a knot and holds
downwards a bough laden with fruit ; the right is lowered and grasps an
object of uncertain shape. It cannot be determined with certainty what
it is that the youth holds in such a peculiar way — possibly a staff. 6
In the corresponding examples also, this hand is always broken awa} 7 . It
is so in a sarcophagus at Ismidt 7 (Nicomedia), and in a fragment not yet
published, which I found in the front garden of the Turkish gymnasium
(Idadie) at Smyrna. Its size is 0'50 metre X 0"7l metre. Fig. 3 shews a
quantity of fragments ; below in the centre is the same youth, with sides
reversed, but with both arms in the same position. Here too the figure stauds
beneath the same rounded pediment in front of the scallop-shell, and between
the same characteristic columns as in the Richmond fragment. The head is
unfortunately broken away. On the sarcophagus at Ismidt the head is
turned to the left, but the figure is placed beneath the same rounded arch.
A parallel may perhaps also be found in the central figure of the Colonna
sarcophagus. s Since, in the fragment B, the left-hand capital is fully
sculptured on the return face, the fragment must belong to one of the angles
of the sarcophagus.

C. — PI. VI. (about 16 in.). This ' statue' closely resembles B and is
almost identical with the two reliefs at Smyrna and Ismidt. The figure
stands resting on the left leg and looks to the left, while the left hand is
raised and hidden under the chlamys. Behind it is the pediment with the
scallop-shell, and on the right the capital of a column.

B. — PI. VII. (ab. 1G in.). This nude youth differs from the others in so
far that he is not standing in full front view, but is turned somewhat to the

2 Miction, Mdawjesd'ArcMologic, xxvi. p. 81. the hoof of ahorse. Mrs. Strong accordingly

3 Munoz, Monumcnti d'Arte, i. 3. thinks the fragment may belong to a lost

4 Mendel, Bull, de Gorr. Hell. xxvi. p. 236. Dioscuius, as on the left of the Sidamara and

5 The same, p. 235. Selefkeh sarcophagi ; or seeing that this basis

6 Mrs. Strong does not consider that the differs in shape from that of the other columns
fragment now plastered up at the bottom on usually found on the long sides of these sarco-
the left of B can belong here. Though the phagi, that it may belong to one of the shorter
foot fairly suits the pose of B's right leg, so it sides (cf. Sidamara, the short side with the
would that of many another figure. Moreover, huntsman, Mon. Piot, ix. Plate XIX. 2).

if the foot is placed correctly in relation to B, 7 Reproduced by Munoz, V Arte, ix. p.

then the base of the column is out of line with 133.

the shaft. ]>y the side of the base may be seen 8 Monumcnti d'Arte, Tav. i. 3, below.

A SARCOPHAGUS OF THE SIDAMARA TYPE 103

right. The curly head is turned to the left in three-quarters profile, and
round it may be seen what appears to be a laurel wreath (?). His movement
is directed to that side as though by stretching out his Left arm, which is
raised under the chlamys, he had to overcome some resistance there. The
identification of this figure as a Dioscurus would be in keeping with the
whole attitude, which we may compare in this respect with the corresponding
figures in our group of sarcophagi. Fig. 4 shews as a parallel example one of
the Dioscuri on the sarcophagus from Sidamara. Baneatb the rounded arch

Fig. 3.— Sarcophagus Fragments ,w Smyrna.

the figure stands in its niche. The position oi the legs is the same. The
treatment is somewhat broader, that is, flatter, but the movement corresponds
exactly, only that the head is more raised. Here, however, the Dioscurus is
reining in his horse — to the left the hind-quartera of the horse are indicated
in low relief; to the right, beside the youth, the forepart of the horse is seen
in a rearing posture, cutting across the column. His lifted forefeet are
carved in bold relief and only united to the body of the sarcophagus by

104 JOSEF STRZYGOWSKI

means of ' puntelli.' These ' puntelli ' also occur on Sir Frederick Cook's
fragment, one below on the pedestal, the other above on the chlamys, in front
of the shaft of the column. Thus one thing only is lacking, and that is the
horse itself, of which no traces exist behind the figure. It is just possible
that it was placed round the corner on the adjoining side of the sarcophagus,
and, as a fact, this figure belongs to an angle of the sarcophagus, for the right-
hand capital is sculptured on its return face. But in that case this sarco-
phagus would differ from all the others of the group. We already find
the Dioscuri on the oldest example in the Riccardi Palace at Florence, 9 then
on the great sarcophagus from Selefkeh in the Imperial Ottoman Museum, 10

Fig. 4. — Dioscurus from the Sidamara Sarcophagus.

and on the Achilles-sarcophagus in the Museum at Konia. 11 It is significant
that Mrs. Strong discovered on the fragment B (PI. V.) the hoof of a horse.

E. — I pass now to two male figures in flowing drapery. Of these the most
important from its motive is illustrated on PI. VIII. Here we have under the
same typical rounded niche a beardless man, in front view. Both hands are
lowered, the left hand holding a roll, the right dragging; one corner of his
mantle to the left. At the same time he looks eagerly to the right; his
curly hair is cut short. Over an ample chiton he wears a mantle which is

9 Strzygowski, Orient oder Horn, p. 52. u Bull, dc Con: Hell. xxvi. p. 225, Fig. 5.

10 The same, p. 47.

A SARCOPHAGUS OF THE SIDAMARA TYPE

105

thrown first over his left shoulder; it is next brought round his right hip and
is then drawn, with a strikingly executed twist, over the wrist of his left
hand. Finally it is gathered below into energetic horizontal folds. Even
this strikingly original figure, conceived, one might say, in the spirit of a
master of Donatello's vigorous individuality, has its analogy on the sarco-
phagus discovered in Asia Minor. The motive of the fold crossing over
the body and over the wrist is to be found also in Fig. 2, the man standing
beside the doorway in the Sidamara sarcophagus. This particular motive,
however, often recurs. It is otherwise with the fold below drawn across
horizontally. To my mind it is curious that this original and peculiar drapery

Fig. 5. — Sarcophagus from Selefkeu at Constantinople.

also should recur in precisely identical form on the great sarcophagus from
Selefkeh at Constantinople. Fig. 5 shews one of the broader sides of this
monument. In the centre we see a nude youth holding a bough in his left
hand, like B of the Richmond sarcophagus ; then on each side a female figure ;
finally, at each of the ends, a draped male figure. Even in the reduced
illustration, and though this figure stands quite at the extreme end, we
cannot fail to recognise that it is identical in every particular — the supporting
leg and the leg at rest, the turn of the head, the position of the arms, the
left hand holding the roll, and last, not least, the two horizontal folds in the
drapery, one above the other.

F. — PI. IX. (17 in.). Thisfigure produces the impression of a Herm : the
upper part of the body is closely compressed, while the mass of drapery
diminishes towards the feet. The arms and hands are entirely covered by a
pallium which falls downwards with a grand vertical sweep. The right arm

106 JOSEF STRZYGOWSKI

under it is raised to the breast ; the left is placed in front of the body ; the
head with its short curly hair looks to the right. As yet I have found no
parallel for the bold scheme of drapery on the Asia Minor sarcophagi.
Finally we have three draped female statues.

G. — The finest of the three (PI. X) might just as well stand on the
Campanile or on Or San Michele at Florence. There it would be appreciated
at its true artistic value, but being on a late antique sarcophagus it
finds no favour. The slender figure stands erect and taper-like ; the turn of
the throat and head do not introduce a sense of movement so much as add to
the impression of dignity and repose. Her right arm rests in a fold of the
mantle ; the left hangs down, holding a roll. Here too the principal charm of
the composition resides in the drapery. The view reproduced in PI. X.
shews clearly how the upper portion of the mantle passes over from the left
to the right shoulder, turns back again below it at an angle, and crosses to the
left hand. The drapery is stretched perpendicularly over the left arm. An
exact replica of this beautiful figure occurs on a sarcophagus at Brussa 1 ' 2 —
without, indeed, the great distinction of style of the Richmond example.
The angle formed by the folds of drapery over the right hip is here mechanic-
ally copied and becomes a mere caricature. The left hand again is lowered,
holding a roll. A similar figure occurs on the Colonna sarcophagus.

H. — PI. XL A figure with dignified and expressive action such as a
Gothic artist might have chosen for Mary in the Annunciation. This
woman turns to the left, but leans back slightly to the right with the upper
part of her body. She grasps the folds of her mantle together in front as if
alarmed, a gesture which suits the serious expression of her face ; her left
hand remains caught in the folds over her breast. For this impressive
creation I know of no parallel among any of the sarcophagus sculptures
hitherto discovered in Asia Minor.

J. — PI. XII. Here we have a woman, in front view, with her head,
over which her veil is drawn, turned to the left. Her right arm emerges
from her short-sleeved chiton and is brought across her breast to her left,
where it rests on the veil ; with her left hand she gathers her veil together in
a bunch which she raises towards her left. This figure is not uncommon —
the pose is such as we find repeated four times on the two longer sides of
the great Selefkeh sarcophagus in the Imperial Ottoman Museum ; the women
on each side of the central figure always hold the ends of their mantle
together with their left hand as in the Richmond example. Fig. 5 shews
that one of these four statues that most nearly resembles our figure. The
woman stands to the right between the central figure and the youth at the
corner. She holds her arm as in the Richmond fragment, across her breast,
but the whole action looks like a weak imitation, whilst the decided power
displayed in the Richmond figure seems to give evidence of an original
creation.

12 L'AHe is. p. 131, Fig. I.

A SARCOPHAGUS OF THE SID AM AR A TYPE 107

The foregoing description and comparison with corresponding sculptures
prove beyond a doubt that the nine Richmond fragments belong to the type
of sarcophagi from Asia Minor to which I first drew attention in Orient
oder Bom, pp. 40 f. My object then was to assign to its proper group in art
a relief with a figure of Christ, purchased in Constantinople for the Berlin
Museum. No pieces of Christian sculpture have been added to the group
then spoken of, but the number of antique examples has meanwhile so much
increased that we may safely say there can scarcely be any of the larger
museums that does not possess a fragment. 13 Therefore when I undertook to
publish the Richmond fragments, it was not to add a new piece to the
sculptures already known, nor because here was a specially well preserved
example such as the sarcophagus from Sidamara, but because, from an artistic
point of view, the fragments belonging to Sir Frederick Cook occupy by far
the most important place and thus can best enable us to enter into the spirit of
that side of the Asia Minor sarcophagi which till now has been neglected by
everyone, namely, the statuary motives. I myself, in 1903, in my book
Orient oder Bom, only went into the subject so far as was necessary in
order to elucidate the fragment in Berlin with the figure of Christ. My
chief object then and afterwards in my work on Mschatta u was to determine
the special style of ornament which decorated the architectural setting and
to investigate its origin. I should like to refer briefly to this before I enter
upon an examination of the actual architectural structure and of the statuary
types.

Only in one of the Richmond fragments (B) were both the columns on
either side of the figure preserved, everywhere else only the column to the
right. It almost seems as if the sarcophagus, perhaps discovered intact, had
been purposely broken into pieces in order that it might be more easily
transported abroad. All these fragments have the same architectonic back-
grounds. The spiral-fluted shaft of the column stands on a Hellenistic
plinth 15 and a peculiar base; the lower moulding projects broadly, in shape
like a plate, while the upper moulding is flat between two broad fillets.
These characteristics are common to all Asia Minor sarcophagi. The capital
is not quite so uniform ; we have here, it seems, an important evolution
which may enable us to fix the date of the sculpture. The capital in the
Richmond fragments spreads out to both sides almost square over the lower
row of acanthus leaves, and displays four remarkably large volutes side by
side. On the oldest example — the marriage sarcophagus in the Palazzo
Riccardi (Fig. G a) — the acanthus is spoon-shaped, and the sarcophagus in the
Colonna garden also deviates somewhat from the Asia Minor type. I here
reproduce one of its capitals (6 b) ; the acanthus leaves lie flat without the
elegant curve of the profile or of the lobes. The effect is obtained more by

13 Cf. the examples brought together by xxvi. pp. 79 f.
Strzygowski, Orient oder Horn, ss. 40 f., and M Jahrbueh der konigh schenKunst-

Byz. Zeitschrift, x. 726, and xv. 419 ; Munoz, tammlun* pp. 205 f.

Kuovo Bull, di Arch. Christ, xi. 81 f., and l5 Cf. GoU elehrte Anzeigen, 1906

L Art,: ix. 132 ; Michon, MiUmget d'Archeol. pp. 911 f.

108

JOSEF STRZYGOWSKI

means of the dark triangular hollows between the little lobes than by the model-
ling of the leaf itself. A similar tendency in the cutting of the leaves is to
be observed in the decoration of several blocks found in the theatre at
Ephesus (6 c). These display on their semicircular face the same vertical
acanthus divided down the centre by the lotus (Fig. 6 c). Here the purpose
of the flat rendering and of the effect of the dark interspaces is still more
striking. Then follows the type which is represented by the Richmond
fragments and the bulk of the Asia Minor sarcophagi (6 d). Hitherto the
foliage had been carved with the chisel ; now it is exclusively worked with the
borer ; the modelling becomes of entirely secondary importance by comparison
with the deeply bored interspaces. The leaf itself in its actual shape really

Tig. 6. — Development of the Sarcophagus-Capitals.

exists no longer ; only an impression of it is produced by the coloristic contrast
between the white, jagged edges of the leaf and the dark background. This
technique, with the intended impressionism of its decoration, is to be found
in the entire group of Asia Minor sarcophagi. Later we find it abandoned
in favour of a method which became common in the Prokonnesos just out-
side Byzantium ; here the chisel completes the work of the borer. This type
is well represented by the capitals of the fragment of the Christ relief at
Berlin (6 c), where the acanthus assumes a novel, thickly jagged, form. 6

16 See on this point my Orient oder Rom, p. 56.

A SARCOPHAGUS OF THE SIDAMARA TYPE

109

To keep to the ornamentation of the Asia Minor sarcophagi in general.
Look at the impost above the capitals (6 d). What has happened to the
Greek egg-moulding in the upper border ? The dark background dominates
the circle in the middle and the strips at the side, and the ornament below
with the trident is meant to represent the ancient Lesbian cymatium ! How-
ever, the whole member, slightly curved at the sides on which these two
strips appear, is scarcely recognisable, owing to its projection, as a part of
the ancient architrave. But as little can it be doubted that the gable-lines
of the pediment with their dog-tooth moulding are the purely decorative
transformation of an antique design. How this transformation, which

^.-A-.-...

Fig. 7.— Relief at Berlin.

probably originated in Mesopotamia, was accomplished, and how the
oriental decorative style of composition in white and black came to sup-
plant the Greek method of modelling the foliage in light and shadow, can
be studied in my work on Mschatta and in an article of mine on the
' Fate of Hellenism in the formative Arts.' 17

In contrast with the important revolution wrought in the rendering of
the forms in ornament where we see the motives handed down from Greek

17 Die Schickaale d<s Hellenismus in der bildenden Ku,isf mNcue JahrbUcher fur das klassische
Alterthum, xv. pp. 19 f.

110

JOSEF STRZYGOWSKI

art conceived in an entirely novel style by means of colour instead of form,
the figures on our sarcophagi keep wholly within the range of a period of
ancient art long previous to the time of their production. I fixed the date of
the Christ relief in Berlin, from which my researches in Orient oder Rom
started, in the third or fourth century, the sarcophagus in the Palazzo
Riccardi is assigned to the Antonine period, and the greater number of the
examples in our group must have been produced in the interval between
these dates. They all belong, therefore, to a period subsequent to the
Christian era, yet the figures on them do not adhere to that picturesque
illusionist tendency in art observable in those Hellenistic reliefs produced
in the centuries about Christ's birth, where the aim was to evoke the
illusion of space, nor do they follow in the steps of the Pergamene and
Rhodian ' Barocco ' of the preceding period. They are typical examples of that

Fig. 8. — Sarcophagus Fragment in the British Museum.

Greek plastic art which flourished before the time of Alexander the Great,
and which sought to solve the problem of the figure in plastic form either
as nudes standing in repose or as highly perfected systems of drapery. It
is the art of the fourth century which again makes its appearance in the
types of statues on our sarcophagi. That is the point on which the
fragments of the Richmond sarcophagus furnish us with such excellent
evidence.

This fact had already struck me in the Christ relief in Berlin. 18 The
Saviour (Fig. 7), in attitude and drapery, shews the type of the ancient
orator, the best known example of which is the statue of Sophocles in

13 Orient oder Rom, p. 59.

A SARCOPHAGUS OF THE SIDAMARA TYPE

the Lateran. The head of the Christ, moreover, closely resembles that of
the Praxitelean Eubuleus. A similar head, somewhat resembling the Eros of
Centocelle and also reproduced in the ancient Christian statuettes of the
Good Shepherd, is shewn in the fragments B and C, the two youths
standing quietly in front view. In position and bearing also they correspond
with that group of figures which we like to associate with the name of
Praxiteles. With the exception of a
few alterations necessitated by the
composition within a niche, B might
be regarded as a copy of the Hermes.
Types like those of the Dioscuri, which
frequently appear on our sarcophagi
as corner figures, trace their origin
back into the fourth and even the fifth
century, and are exemplified in the
fragment C. It is not impossible that
the Richmond sarcophagus, when it
was still perfect, displayed as the
principal figure on one of its longer
sides a seated statue, such as we find
on the sarcophagi of Sidamara ami
Selefkeh. The British Museum pos-
sesses a fragment of the same kind
(Fig. S). We have here a composition
in flattened relief showing a bearded
man reading from a roll. In front of
him stands a Muse with the tragic
mask. Her head-dress points to the
Roman period. The type itself, how-
ever, again belongs to pre-Alexandrian
art and has it origin in the reliefs on
Attic tombs in the style of Pheidias.

The chief evidence for the purely
Greek origin of the types of statues
on the Richmond fragments is fur-
nished, in my opinion, by the draped
figures. Original works in the style
of E energetically clutching the folds
of his garment like some prophet
of Donatello's are incredible in the
Roman period. This figure evinces so
much individual creative power that

it can only belong to a period of unusual activity in the domain of
fonn -problems in statuary. A similar movement to that of the tolls in the
herm-like figure F — the end of the mantle drawn from the right shoulder
straight across the breast and over the left hand — is to be found on one

Fig. 9.— Muse feom the Maniiman
Basis.

112

JOSEF STRZYGOWSKI

of the Muses of Praxiteles on the basis from Mantinea. Notice on one
of these reliefs (Fig. II, from the slab with the three Muses standing) what
economy of line prevails in the arrangement of the principal folds round
the breast and you will then understand why I venture to assign to the

Fig. 10. — The 'Matron of Heiicilankum.' (Dresden.)

beautiful draped figure G, as regards its type, a place near Praxiteles. The
so-called Matron of Herculaneum in the Dresden Albertinum confirms me in
this view (Fig. 10). In this single statue, as in Sir F. Cook's fragment,
the chief form-value consists in the arrangement of the folds on the left

A SARCOPHAGUS OF THE SIDAMARA TYPE. 113

breast. They seem to be drawn tight between the shoulder, the raised right
hand and the left, which is held down. The position of the head and legs
in the Richmond fragment has been changed for the sake of that
correspondence between neighbouring pairs of figures which is usual in the
Asia Minor sarcophagi. The veil, too, is absent, as the pileus frequently is
in the case of the Dioscuri. Amelung 19 considers this type of the statue
from Herculaneum to have had its origin in the school of Praxiteles, and
P. Hermann, who has daily opportunity of studying this grand work in the
Albertinum, confirms this opinion in a letter : ' The Dresden statue is a
faithful copy of a sculptured original of the fourth century, most probably of
the circle of Praxiteles. Head and body belong inseparably to each other
and form a complete artistic whole. In the Roman period this Greek type
sometimes served for portrait statues, and would be given a portrait head in
the place of the ideal head belonging to it. This, however, is not the case
with the example from Herculaneum.'

This comparison brings us back once more into the ' milieu ' to which
the figures of this sarcophagus belong — to the time when artists subsisted
on their Greek heritage from the pre- Alexandrian period, and were actively
employed in copying ancient types. Presumably, therefore, the types of the
two draped female figures H and J are likewise not new creations by an
Asia Minor sculptor of the Christian era. They go back to a school which is
represented by an original ; the sarcophagus, namely, with the ' Mourners '
(Les Pleweuses), discovered at Sidon. 20 There, too, we find the same division
of the walls of the sarcophagus into separate niches in which are placed, each
one alike and by themselves, the separate mourners, as in our Asia Minor
sarcophagi .

The sarcophagus of the Mourning Women was found at Sidon in Syria ;
the details of its sculptures leave no doubt of its connection with the art of
southern Asia Minor. It now remains to prove that the Asia Minor
sarcophagi also belong to this school of plastic art, and depend from a centre
of which till now we knew very little, namely Antioch. For to the sphere of
influence of this Syrian metropolis belongs also the region on this side of the
Taurus whence the art tendency noticeable in the Richmond fragments may
have travelled to the west of Asia Minor just as well as to Macedonia, Greece,
Italy and Rome. For the present nothing can be determined with certainty,
but it is my firm conviction that the Asia Minor type of sarcophagus had its
origin neither at Ephesus nor in any other district of western Asia Minor,
neither in Greece nor Rome, but in the angle which lay nearest to Meso-
potamia, and had Antioch as centre of culture. In proof of this I should like
to bring forward certain considerations.

19 Die Basis des Praxiteles aus MatUineia, pp. (' Alterthum '), pp. 280 f.; cf. aann,

26 f. ; S. Reinach, however, ascribes it t" Ly- Kunstg schichte, i. p. 154 ; E. Strong in
sippus (/.'■ o. Arch. 1900, ii. pp. 380 1'.), likewise Ci u; 1901, pp. 1S7 f.

Collignon, 'Lysippe' (L Hamdy Bey and Tli. Reinach, Les

p. 21, p. Ss . A middle view is taken by Sarcoph
Michaelis in Springer's . i-

U.S. — VOL. XXVII. I

114

JOSEF STRZYGOWSKI

In favour of the neighbourhood of Mesopotamia is the composition of
the ornament by means of dark interspaces, while in proof of the Antiochene
district of Asia Minor I think I can appeal to another characteristic to which
I have not yet called attention ; namely, the peculiar architectural frame
in which the Attic figures are placed. Their shoulders are surrounded
by scallop-shells as by an arch. But these shells are simply a shape borrowed
from Nature for an originally purely artistic form — the niche. Where is the
home of the wall niche that ends at the top in a flattened quarter sphere ? It
is unknown either to Egypt or Greece, those two great master-powers of wood

Fig. 11. — Xiciie in Cemetery of the Tulvnids, near Cairo.

and stone building. A priori, therefore, it might well have originated in the-
East ; and this is supported by the fact that the art of Islam, in the final form
in which it penetrated from Bagdad into the West, makes such an extensive
use of the niche that the entrance of all secular and sacred Mohammedan
buildings develops into the typical niche form of the ' Sublime Porte/ and
also the Mihrab, which takes the direction of the national sanctuary at Mecca,
is given in this same shape. Even the shell is to be found in one of the
oldest examples ; Fig. 11 shews this Islamic form. I found this shell-niche in
the year 1895 in the cemetery of the Tulunids, which lies to the south of

A SARCOPHAGUS OF THE SI DAM AR A TYPE. 115

Cairo outside Bab-el-Karafa, and of which Makrisi, as early as the year 142U,
reports that there were many oratories there, into which holy men were wont
to retire. 21 The Mihrab of Imam Schaffai, situated near the mausoleum, the
chief holy place of this district, was probably the last remains of one of these
oratories (it has since disappeared). In my photograph the shell can be
plainly seen ; it was rendered in stucco, and the ornaments in the spandrils,
which complete the pointed niche in a square, are carried out in the same
material. This frame motive itself as well as the tendril-work is of typically
Persian origin. Evidences for this can be found in my work on Mschatta, in
another on the miniature painting of Lower Armenia, 2 ' 2 and above all in the
stucco decoration in the mosques at Cairo 23 brought there from Persia by the
Tulunids, the Fatimids and the Ayyubids. I merely make use here of the
Mihrab, which repeats the type well, in order to shew the reader the
further development of the shell-niche on the ancient soil of Mesopotamia,
whence it probably had also made its way into the ancient architecture of
Asia Minor and of Syria. There is the original home of the brick wall u
divided on the outside by flat, on the inside by rounded niches. This style
of wall construction, translated into stone, first makes its appearance in
the great temple buildings and Nymphaea of Syria and Asia Minor. It is
probable, therefore, that the group of sarcophagi which developed this motive
as its type belongs also to this group.

It has never yet been noticed at all that the key to the explanation of
how and where this style of sarcophagi could have developed is supplied by
the Christian ivory carvings. At a time when the foundations of the study
of Christian antiquities is about to be laid, unfortunately on a philological
basis, 25 I am glad to be able to shew what very surprising disclosures plastic
art alone can lead to in this direction. It is significant of the methods of
classical archaeology that it has got into this channel, not with the help of
my labours, but just now in the footsteps of Literature. It still clings more
than one would think to letters instead of opening its eyes to the forms and
figures of painting and sculpture.

Fig. 12 shews the front of the celebrated throne of St. Maximian at
Ravenna, In the centre stands John the Baptist in front view ; the weight
of the figure is on the left leg and the right is at ease. The saint raises his
right hand to bless in the Greek fashion, and holds a disk with the lamb in
his right. He is flanked by two evangelists on each side, who each turn
towards the central figure ; their gestures are varied, but all carry their
symbol, a volume, which they hold in their left arm. Examine closely the
motives of the splendid drapery, which are varied in each figure, and the
richness of the folds with their individual arrangement, and you will come to
the conclusion that these five figures disposed round a central figure are in

21 Wiistenfeld, Oeschichte dcr Koptcn, p. SO, - 3 On this point sec Franz Pasha, Cairo.

13. 4. '- 4 Strzygowaki, KUinasien ein Keuland dcr

'-'- System, alphab. Hauptkatalog der K. Unic. Kunstgcschichtc, p. 38.
Bibl. -.« Tubingen, xiii. ' Yerzeichnis der ar- Von Sybel, Christliche Anlike, 1906.

in. nischen Handschriften,' Tubingen, 1907.

I 2

U6 JOSEF STRZYGOWSKI

attitude and drapery really nothing but the longer side of one of our Asia
Minor sarcophagi (cf. Fig. 5). It is true that in place of the monolithic
stonework which permitted the architecture to form a continuous frame round
the figures, the subtle technique of incrustation has been introduced into
ivory carving, and to suit this technique three rectangular frames are joined
together. Between these, however, narrow pieces are inserted just as in our
sarcophagi. And hand in hand with the introduction of framework came
another innovation. The niche architecture on the sarcophagi, with its
projecting entablature, had brought about as a necessary consequence that

Fig.

Tiiuone of St. Maximian, Ravenna.

figures in niches should alternate with those which stand in front of the
straight architrave connecting the niches. The ivory carver composed panel
by panel, and therefore executed the niche motive singly for each figure— but
yet he could not emancipate himself from the customary arrangement of
broad and narrow spaces. This scheme, which is inexplicable for ivory carving-
considered by itself, affords the clearest proof that the sculptor of the
pulpit of St. Maximian is closely connected with the art of the Asia Minor
sarcophagi.

A SARCOPHAGUS OF THE 8IDAMARA TYPE.

117

Now the throne of St. Maximum, as I have shewn elsewhere, 26 is of
Syrian origin. Its sharply cut tendril-work decoration, with dark interstices,
points to this. I thought of Antioch itself as the place where it was pro-
duced, and can now support that assumption by its relationship to the Asia
Minor sarcophagi on the one hand, and on the other to a monument which
will also disclose to us most remarkable evidence with regard to the origin
of the whole group and the question of date. I allude to the beautiful ivory
diptych in the British Museum carved with the figure of an archangel and
bearing the legend AEXOY riAPONTA KAI MAOCUN THN A°TIAN.
(Fig. 13.) The decoration in this case leads
us to presume that it is of Syrian origin.

Place it now beside the Asia Minor sarcophagi

and the throne of St. Maximian, and we shall

find that it has points of contact with both.

The arrangement of a single figure in a niche

containing a shell is the same as on the sarco-
phagi. Closely related to the statues of the

throne is the beautiful bold motive below — the

roll drawn round the hips. There we see one

end of the mantle laid broadly across the body

to hide the lap. The same trait occurs in the

two apostles who stand nearest to John on the

pulpit. The beardless apostle to the right

shews almost exactly the same motive as the

archangel ; in the bearded one to the left, on

the other hand, this mantle is drawn across the

body down to the left knee. While there is

no doubt that the throne is later than the

sarcophagus, whose scheme of composition it

evidently presupposes, the archangel diptych

exhibits a motive which has an appearance of

great antiquity.

Look back to the sarcophagi. The figures

there stand between the bases of the columns

on the lower border. The arrangement on

the throne is more individual, because there

attention has been paid to the profile-edge of

the three large panels. Now the form of the

ground in the London ivory diptych is quite

unique. Between the deep fluted pedestals six-
steps lead right up to the height of the bases of the pillars. What did
the sculptor mean by this unsuitable motive ? He was thereby only getting
involved in contradictions, for as a sculptor of merit, and for the sake of the
representative character of his figure, it was important for him to place his

Fig. 13. — Leaf ok Ivory Piiiwh
in the British Museum.

28 Jahrbuch der preussiscken Kunstsammlungen, 1904, p. 299, 'Mschatta.

118 JOSEF STRZYGOWSK1

archangel as near as possible to the surface of the panel, that is, immediately
in front of the spectator. But the steps required that the archangel should
appear on the platform above, that is, standing in the space far behind the
pillars, pushed back, in fact, into the room. Instead of that he presses forward
right in front of the shafts of the columns, and his arms and even his wings
hide both shaft and capital. As a consequence the lower part of the body
would have to be represented retreating towards the background. The sculptor
partly gets himself out of this dilemma in a most naive manner. He cannot
quite bring the feet into the plane of the composition — the figure must
remain upon the platform — that is part of his fixed idea. So he lets it stand
up on the platform with its heels, while the soles are stepping down on three
steps at once. This exaggeration of the feet seems to him preferable to
giving up the whole motive.

Why this stiff-necked obstinacy ? Riegl, in characteristic fashion, finds
a definite artistic intention hidden in the motive. He thinks the artist
•'conscientiously avoided representing a definite momentary kind of standing
on the steps, and endeavoured rather to set this act of standing before the
eye of the spectator as objective type and in order to characterise the feet,
by means of their upper surface, as giving the effect of depth.' 27 I have often
enough taken my stand against this sort of theorising, and may, I think, in
this case declare for once emphatically how mistaken such well-meant
explanations are, when they so entirely neglect historical facts, as Riegl does
in this instance.

Precisely as the curious arrangement of the five figures on the front of
the throne of St. Maximian betrays in respect of the Asia Minor sarcophagi
an atavism manifest in certain inconsistencies, such as the alternation of broad
and narrow fields, so here the sculptor of the London ivory diptych does not
advance with a will towards the discovery of new motives, but shews himself
retrograde in his weak adherence to traditional ideas. The six steps between
the pedestals in front leading to the background of the relief are not his
invention, but go back, together with the motive of the doorway to which they
lead, to presupposed facts, the demonstration of which must for the time being
be sought for at Pompeii. Suppose a theologian were making researches in
the houses there in order to elucidate the motive of the steps, and came in
the Casa di Marco Lucrezio, for example, to the steps which lead up from a
fountain to a statue standing in an arched niche, he might well imagine that
the artist of the ivory relief, by analogy with the cascade and its flight of steps,
had wished to convey that the archangel, like the water which we suppose to
flow down the steps, was the bringer of life. As a matter of fact the Good
Shepherd and Daniel were placed in this fashion by Constantine the Great

27 Alois Riegl, Spatromische Kunstindustric, Stehen iiber den Stufen an sich, als objektiven

p. 122: the artist 'hat es geflissentlich ver- Typus, dem Beschauer vov Augeu zu fiihren unci

mieden, eine bestimmte momentane Art des die Fiisse vermittels der Obersicht als tief-

Stehens auf den Stufen darzustellen, da sein ranmerftillend zu charakterisieren.'
Streben vielmehr darauf gerichtet war, das

A SARCOPHAGUS OF THE SIDAMARA TYPE.

119

over the fountains in the centre of the squares of Constantinople. 28 By
similar combinations the Indian figure of the boy on the water-mound on which
animals are feeding and drinking received a Christian interpretation. 20
Another might discover the flight of six steps of our relief in the six steps that
lead up to the temple of Isis at Pompeii and its vestibule. As the cultus
image in its day was shewn with this architectural approach, so the
sculptor (we might say) introduced his Christian archangel, steps and facade
with columns going back therefore to antique temples raised on a podium.

In reality the key to the origin of our motive lies much further away
from the path which is usually followed in tracing the motives of Christian
art, and many a Christian archaeologist
will probably tear his hair out with
horror, when I propose to prove that
the archangel in the London relief has
been placed here like a real actor on
the stage. Perhaps some may be more
inclined to reflect seriously on what I
am about to say when I mention that
lately a theologian has tried to point
out, in the Archiv fur Rcligionswisscn-
schaft, ix. 3G5 f., that the sculptured
screen (the ikonostasis) of the Ortho-
dox Church had its origin in the Pro-
skenion of the ancient theatre, and
that characteristic features of the
Liturgy, such as the elaohoi, were none
other than the acts of the Hellenic
drama, so that when we Westerns
reproach the Orthodox Church for its
theatrical services our reproach is in
the truest sense justified. Karl Holl,
as I shall point out, was quite right
in making these assertions.

For the archangel on the steps a
convincing analogy is to be found at
Pompeii in the wall paintings of the
fourth style only. There one often sees
(best in the stucco decoration of a wall in the Stabian Thermae and in some
paintings in relief which are now in the Bronze Rooms of the Naples Museum)
figures between columns, represented standing in a doorway to which steps
lead up. The annexed example is from the so-called Palaestra (Reg. VIII.
ins. 2, No. 23), Fig. 14. 30 We see between the projecting side walls with their

Fig

11. — Portion of Pompeian "Wall-
Painting.

- 8 Cf. Bomiseht Quarlalschrift, iv. 102.
29 Such groups in ivory are to be found in
nearly all museums.

•" This illustration and the following are

alter von L'ulie. Die ESmitehe ' scenae frons '
in <!■ n schen Wandbildern, Berlin,

1906.

120

JOSEF STRZYGOWSKI

columns five steps and then two more leading up to the nude youth who
appears above in the doorway. The sculptor of the London diptych also
imagined the side walls to recede in a similar way. He would otherwise not
have placed the archangel's sceptre on the pedestal to the right, and by thus
correcting the columns have moved them back somewhat into space. The
number of the steps, which, as in the Pompeian picture, get smaller to the top,
also almost coincides. The fact is that for such flights of door-steps in the
Campanian wall paintings five steps on an average are used.

I do not mean to assert, in quoting this analogy, that the ivory sculptor
had actually copied a Pompeian picture or any antique painting at all. It may
be fairly clearly established in this instance that between Pompeii and the
diptych there is a third connecting link — the ancient stage. This can be
proved with the help of those monuments which are most closely allied to the

Fig. 15. — Reconstruction of Pompeian Stage Facade.

archangel relief, such as the front of the throne of St. Maximian and the
Asia Minor sarcophagi. The five figures side by side are distinctive of them.
These figures are placed — on the throne and on the sarcophagi — within and
between the three pairs of columns which are connected either by a rounded
arch or by a pediment (Fig. 5), and which, as the archangel relief with its steps
shews, were meant to indicate doors. But what are these three doors united
into a whole by a projecting entablature ? That is the actual division of the
stage wall which Holl has accepted for the ikonostasis of the Greek Church
and which — a fact I have not yet mentioned — Puchstein has assumed to be
the model for the Pompeian wall paintings of the fourth style mentioned
above.

Lately von Cube, at the suggestion of Puchstein, undertook, by con-
fronting what is preserved of the extant ruins in the theatres themselves with

A SARCOPHAGUS OF THE SIDAMARA TYPE.

121

what can be made out clearly from the wall paintings, to reconstruct these
stage walls. I give here (Fig. 15) an example done from the very fanciful
wall picture introduced into the upper part of the architecture in the
triclinium of a Pompeian house (Reg. I. ins. 3, No. 25). We see here the
three doors with five steps each ; in the centre the Aula regia, to the side
the Hospitalia. They are flanked by columns on pedestals, and where the
pairs of columns come close to each other statues stand in the narrow inter-
space. If these are restored from the originals or from the ivory tablet in
the case of the doors also, we shall have first the wall painting itself (Fig. 16),
which is the foundation of von Cube's reconstruction, and then the long
facade of an Asia Minor sarcophagus (Fig. 5). For it is obvious that if we
reconstruct the upper part of this architecture, not entirely from the painting
as von Cube has done, but according to a reasonable architectural point of

Fig. 16. — Pompeian Wall Painting.

view, then arch and pediment would come over the doors and not between
them. It will be well, therefore, when reconstructing the stage walls of
ancient theatres in the future, to take into consideration the long sides of the
Asia Minor sarcophagi. But if the question arises as to where this theatre
architecture penetrated into painting and sculpture, and if it is urged that
this could only happen in a great city, then probably the innovation should
be referred to Antioch rather than to Rome. It is from Antioch that
the fourth style of Pompeian wall painting 31 and the type of the Asia Minor
sarcophagi came; from thence also, or from one of the islands lying off the
Syrian coast of Asia Minor, come the Ravenna sarcophagi, which reproduce

31 Cf. Strzygowski, Gottingiscfie gelehrte Ameiuen, 1906, pp. 910 f.

122 A SARCOPHAGUS OF THE SIDAMARA TYPE.

the type of the theatre facade in its latest phase, posterior to the throne of
St. Maximian — the type, namely, with the five arcades of perfectly even
Avidth. Finally, from Antioch comes also the ivory throne itself and the
archangel diptych in the British Museum, in which the motive of the theatre
steps has been so strikingly preserved. Just as in Japan there were times
when painting and sculpture remained entirely under the influence of the
theatre, so also in ancient art. It appears that this was the case at Antioch.
We find the fashion had penetrated to Italy with the fourth style in the time
of Nero. In the period of the Antonines it dominates sarcophagus sculpture
in the central district of the eastern Mediterranean ; in the archangel relief,
the Ravenna throne, and the Christ relief at Berlin it encroaches on
Christian art, and is destined afterwards to celebrate its final triumph in the
Ravenna sarcophagi and to live on unrecognised up to the present day in the
ikonostasis of the Greek Church.

To conclude, I return again to the Richmond fragments. They belong
in every particular, in the decoration executed with the borer, and in the
beauty of form of the statues and the strict adherence to the architecture of
the theatre wall, to the best specimens we possess of the Asia Minor school
of sculpture which had its starting point in Antioch. Whether these
sarcophagi were produced at Tarsus, as Sir William Ramsay 32 thinks, or
whether, as I supposed, the marble points to the Prokonnesos 33 (therefore to
the ancient Cyzicus), and whether the Richmond fragments come, as
Th. Reinach 3i concluded in the case of the Sidamara sarcophagus, from the
mountains north of the Taurus, or — as the export of Greek marble sarcophagi
to Ravenna led me to suppose — from one of the islands lying off the coast of
Asia Minor, are points the decision of which is reserved to the researches of
the future.

The sample sent to me by Mrs. Strong shews that the marble of the
Richmond sarcophagus comes from Greek quarries. It is grey in colour and
crystalline throughout. I also have samples of the sarcophagi from
Selefkeh in Constantinople, of the Christ-relief in Berlin, of one of the
sarcophagus fragments in the Louvre, and of the sarcophagus in the Colonna
garden. The last two shew pure white crystals. On the other hand, the
Richmond sample comes near in its grey colour to the examples from
Selefkeh and the Christ-relief from Constantinople in Berlin.

Josef Strzygowski.

(Translated from the author's MS. by Mrs. Arthur S'rong.)

32 Revue des fitudes ancienncs, 1901, p. 358 ; 33 Orient odcr Bom, p. 54.

Studies in the History of the Eastern Provinces, 3i Monuments Piot, ix. p. 7.

p. 60.

THE 'LIST OF THALASSOCRACIES' IX EUSEBIUS : A REPLY.

Mr. Fotheringham's paper, if I understand it rightly, divides itself
into three parts. He corrects certain errors in my tabular statement of
the evidence, and in my commentary on it; he criticizes my suggestion
that the 'List of Thalassocracies ' may represent a fifth -century docu-
ment; and he reconstructs from materials comprised wholly in the existing
texts of Eusebius' work an outline of the Eusebian view of the 'List,'
as he understands it; the gist of which is that he thinks that Eusebius
not merely incorporated in his Canoncs the names of all the states contained in
the ' List,' but also intended to space the thalassocracies according to the
numerals contained in the Excerpt. From this he infers that, where the
numerals in the Excerpt are lost, they may safely be restored from the
intervals indicated in the Canoncs ; and from this, finally, (1) that the Excerpt
was not mutilated when Eusebius incorporated it in his Chronographia,
and (2) that it represented, in its missing section at all events, the same
chronological scheme as underlies the rest of Eusebius' work, and not,
as I had been led to suggest, a different, earlier, and more accurate
system.

With the permission of the editors of the Journal, I submit a brief note-
on each of these points, in the order indicated above.

(1) Sundry errors of transcription and reckoning.

I hope that my use of the word 'Eusebius' to denote 'the Armenian
version of the Chronici Canoncs ' and of the word ' Chronicon ' for the
' Chro7wgrapMa' has not inconvenienced anyone besides Mr. Fotheringham :
still more that he has not been misled already by the occurrence of the words
Chronicorum Liber I. instead of Chronographia at the top of p. 226 in Sehoene's
edition, on which the Excerpt from Diodorus is printed.

In the matter of the relative value of the Armenian version and Jerome's
version of the Canoncs, I should gladly bow to Mr. Fotheringham's great
knowledge of the Eusebian texts, were it not that on Mr. Fotheringham's
own showing, Jerome's version is more erratic in its support of Mr. Fother-
ingham's theory than even the Armenian version is, and requires even
more ruthless emendation before it can be taken to represent a text
of the Canoncs which shows signs of having utilized the numerals of the
Excerpt.

124 JOHN L. MYRES

For Mr. Fotkeringham's correction of my errors of addition and
subtraction in columns H and J, I have to thank him very heartily; and
I am glad to be able to agree with him that they do not affect my argument.
In columns E and G, I have copied the wrong numeral between
places XV and XVI ; and the corresponding correction which must be
made in my article, on p. 92, 1. 9, invalidates the argument contained in
11. 9-11.

The numeral 45, which I attributed to the Phoenicians in column F,
rests, as Mr. Fotheringham says, on one manuscript only ; but it is defended
by Mr. Fotheringham himself as a genuine piece of the Eusebian text. The
only difference between us, if I understand his criticism rightly, is that
he does not think that its preservation in MS. ' F ' is sufficient to assign it to
Jerome's version of that text.

Meanwhile I am glad to find that Mr. Fotheringham's experience
of MS. ' F ' of Jerome leads him to prefer 32 to 23 as the Eusebian numeral of
the Cypriotes in place VI : for in the event of my interpretation of places
VI- VIII being upheld (pp. 121 and 122 n.) the date 742 B.C. has several
advantages over 732 B.C. It has, however, of course, the superficial
disadvantage of having approximately the duration of a conventional
' generation of men.'

The errors which Mr. Fotheringham has discovered in my summary
of the evidence of Syncellus, in column D, are more serious. After some
trouble, I have discovered how I came to make them, but that concerns
no one but myself, and does not mitigate the blunder. I have accordingly
to cancel the sentence on p. 90, 11. 10-12 of my commentary on column D,
and to substitute the words ' But in places VI -XI inclusive, and in place
XIII, he omits both the names and the numerals.' I must also cancel
the statement on p. 127 that 'Syncellus is silent' as to the duration of
the seapower of Lydia : also on p. 92, 1. 17, for' like' should be read
' unlike.' But the other references to Syncellus in my article are, I believe,
unaffected.

I have also, obviously, to modify my inference that Syncellus was using
the Chrono gravida rather than the Canones; and to lay less stress on my
suggestion that the text of Diodorus was already mutilated when it came to
the knowledge of Eusebius. My argument, however, from the misreading
preserved by Pliny (p. 105) remains untouched; and the question in what
way the missing numerals are to be supplied is not affected one way or the
other.

But while admitting that Mr. Fotheringham has made out a strong case
against my suggestion (p. 92, n. 21) that Syncellus was relying on the
Chronographia rather than on the Canones, I do not see that his conclusion
follows necessarily ; at all events it does not preclude the idea that Syncellus
had the Chronograpihia as well as Canones before him, and was in fact using
both. If his copy of the Canones had contained a mention of the Naxians, it
is difficult to see why he should have made his mistake of ten years in regard
to the Lacedaemonian seapower; and in face of this mistake, it is difficult to

'LIST OF THALASSOCRACIES' IN EUSEBIUS : A REPLY. 125

aigue that Syncellus' copy of the Canones inentioned the Naxians at a point
where all known versions of the Canones omit them. On the other hand the
circumstance that, lower down, Syncellus does mention the Naxians, and
gives them their proper numeral 10, seems to me to suggest that at this
point at least he is using the Chronographia, at all events as a supplementary
authority.

The only reason why this question, whether Syncellus had access to the
Chronographia or not, was worth further discussion, is this. It is only by the
assumption that Syncellus used the Canones exclusively, that Mr. Fothering-
ham is able to make good his generalization that ' each thalassocracy is
found in at least one of the three documents from which the Eusebian text
must be reconstituted.' This assumption of course he can only demonstrate
by showing that Syncellus was not indebted, on any given occasion, to any
other source such as the Chronographia. But the considerations which I have
stated seem to show not merely that Syncellus had access to the Chronographia
but also that his copy of the Canones credited the ten Naxian years to the
Lacedaemonians and consequently did not contain the name of the Naxians at
the point where he puts them : and if this was so the Naxians would seem to
offer a clear case of a thalassocrat state which was not mentioned in any of
the three documents in question.

(2) The List in the Excerpt, and the Thalassocracies of Thucydidcs.

I do not quite understand what points Mr. Fotheringham means to
indicate in reply to my suggestion of a fifth-century date for the list, as
showing that the retrospect of seapower given by Thucydides disagrees with
the evidence of the list. Thucydides does not as a matter of fact give any
' dates ' at all ; and the circumstance that at a time which Eusebius (not
Thucydides, nor Herodotus, nor any early writer) dates as falling within a
' Carian ' seapower, a Corinthian built ships for Sauios, would only prove
anything, if Mr. Fotheringham were prepared to maintain that during a
* Carian' seapower neither Corinth nor Samos was allowed to have ships at
all. To admit the existence of an 'Ionian' seapower ' later' than the genera-
tion of Ameinocles does not seem necessarily to exclude an earlier one; if
only because there was more than one state in ' Ionia.' Moreover the very
circumstance that Thucydides, when he exceeds the data of the Excerpt, does
so only by including Western seapowers, goes far to explain both the omission
of such a state as Corinth from the List, and also the preponderance of
Levantine states in it. The List is clearly a sequence of Eastern Mediterran-
ean seapowers: and consequently not only Corinth but Corcyra. Cumae,
Syracuse, and Tarenturn, as well as the Carthaginians and Tyrrhenians, are
absent naturally. The reported 'seapower of Sinope' is another case in point,
for Pontus is excluded likewise. Also if the List were really a document of
the period ( if the Delian League, and compiled under its influence, as 1 have
suggested, no 'tendency' would be more natural than disparagement of the
seapower of Corinth : a motive too which would have no obvious explanation
except under fifth-century circumstanc -

126 JOHN L. MYRES

The West, in early Greece, as in Thucy elides' own time, was a kind of
' high seas,' common to all. Even in 445 B.C., the very treaty which recog-
nized formally the existence of Leagues, which in the Aegean and on its
shores were practically inclusive, left an open field in the West, where Athens
and her rivals might compete freely for adherents. In Greek thought, the
first ' thalassocrat ' in the West was probably Dionysius of Syracuse.

(3) The Eusebian ' Canones ' as material for the restoration of the List in
the Excerpt.

Mr. Fotheringham's ingenious reconstruction of the text of the Eusebian
Canones, so far as they are concerned with the thalassocracies, is a definite
and valuable contribution to the study of the Eusebian text ; and clears up
many points which are obscure to those who are unfamiliar with its history.
His object clearly is to show that, even in the present unhappy state of the
Canones, enough similarity remains between their allusions to thalassocracies
on the one side, and the names and numerals preserved in the Excerpt on the
other, to justify the hypothesis that where the numerals are lost in the
Excerpt they can be restored from the allusions in the Canones ; that conse-
quently there is no need to go outside the Eusebian text for materials for
such a restoration as I have attempted ; and that the List as restored by Mr.
Fotheringham's method agrees with the Eusebian chronology, instead of sup-
plying materials, as I had suggested, for its correction.

A question of method confronts us here at once. To reconstruct the
text of the Canones on the hypothesis that its true intervals would be repre-
sented by the numerals of the Excerpt, and then to reconstruct the Excerpt
on the hypothesis that the lost numerals would be represented by the
intervals in the Canones, seems to me to be an argument in a circle.

On Mr. Fotheringham's own showing, aberrations of a year or two must
be assumed to exist in the Eusebian dates almost throughout ; and he makes
out a good case for textual aberrations of as much as ten years. Two points
therefore arise. First, how, in the absence of such a clue as is afforded by
the Excerpt in the Chronographia, would it be possible even for Mr.
Fotheringham to divine whether he ought to allow, in the case of any
given event, for a rise of as much as ten years, or for a fall of a similar
amount ? With the clue at hand, it is comparatively easy to suggest how
this or that textual obstacle may have produced the actual discrepancy ; but
what evidence is there even for the existence of a discrepancy, in the cases
where the Excerpt is defective ?

My meaning will perhaps be clearer if we take an actual instance. That
Eusebius had some reason for putting down events at the point where they
occurred, in the original text of his Canones, is in any case probable. That he
should have thought that there was an Egyptian thalassocracy from 796 B.C.
to 753 B.C., as Mr. Fotheringham suggests, is also probable, both for the
reasons which Mr. Fotheringham gives, and for those which were given long
ago by Dr. Goodwin from non-Eusebian evidence as to Hellenic ideas about
Egypt. But nothing of this seems to me to prove anything as to the lost

'LIST OF THALASSOCRACIES' IN EUSEBIUS: A REPLY. 127

Egyptian numeral in the Excerpt ; and unless it is possible to show that the
lost numeral agreed with Eusebius' calculation of the reign of Bocchoris, my
suggestion that Eusebius either neglected, or did not know, the lost numeral,
remains unanswered. On the other hand, even if it were possible to show
that the Eusebian date for the Egyptian thalassocracy was based upon the
lost numeral in the Excerpt, and that this numeral was 43, as Mr.
Fotheringham conjectures, all that would be proved would be that in
regard to Egypt the compiler of the List was working on some lost
Greek tradition about Egypt: nothing would have been gained either in
proof, or in disproof, of my suggestion that the numerals in places VI-VII
correspond with certain actual sequences in Oriental history. Meanwhile,
even the consistency of the Eusebian data can only be tested in cases
where there is something to compare ; and in the case of the lost numerals
this something does not exist.

It was not, however, the consistency of the Eusebian data with them-
selves that I was mainly concerned to discuss in my article ; but rather the
accuracy, or the veracity, of the Eusebian chronology in general, when com-
pared with that fragment of pre-Eusebian chronology— whatever its date —
which is preserved in our mutilated ' List.' When every numeral in a series
of seventeen items is liable even at the hands of its defenders to ruthless
conjectural revision; when the limits of such revision range in individual
instances from as little as two to as much as ten years; when the effect of
these errors is cumulative as we recede from the starting-point, and when the
starting-point of the list itself is admittedly five years wrong, it is per-
missible to doubt whether there can be any very positive evidence that
the original Eusebian dates conformed at all closely to the numerals of
the Excerpt or even that the general spacing of the Thalassocracies in the
Ganones stands in any very close relation to whatever chronological scheme
the Excerpt may have embodied when it was entire.

That the thalassocracy-entries in the Canoncs formed a separate system
by themselves is of course proved directly by the existence of the Excerpt,
and is independent of any considerations derived from the use of red or black
ink. What even Jerome meant by the use of red or black ink is far from
clear; and what Eusebius meant by those features in his Canoncs, which
suggested the use of red or black ink to Jerome, is obscurer still. Least of all
is it clear from Mr. Fotheringham's discussion, or from any other sources with
which I am acquainted, whether it was the Thalassocracy list which (with
other such lists) formed the groundwork of chronology upon which the other
events were spread about in their probable order, or whether the thalassocracy-
entries (whether made in the margin or interpolated) represented rather a
late phase of the compilation, and only found place in it at all, when the
other Eusebian data had begun to give cumulative proof that any given state
could only be credited with seapower between such and such dates. If the
former theory be accepted, then either Eusebius did his work very badly, or
his text needs re-writing, in the way Mr. Fotheringham has proposed, till it
conforms to the data of which it is compounded; if the latter, it is a matter

128 JOHN L. MYRES

of indifference what the text of the Canones may have contained, for ex
hypotkesi the numerals of the Excerpt were only retained when they
happened to fit the Eusebian theory of history.

As to the Armenian version, meanwhile, Mr. Fotheringham is brought
to the same conclusion as myself, that whatever the system may have been
which these entries were intended to embody, they are scattered about in a
manner which he rightly attributes to the ' general confusion of the
chronology of this version.'

As Mr. Fotheringham's argument for the consistency of the Eusebian
data has led him to discuss certain other Eusebian allusions to states which
had seapower, I may perhaps be permitted to illustrate from these allusions
the discrepancy between Eusebian chronology in general and our present
knowledge either of fifth-century Greek chronology, or of the actual course of
events as determined from non-Hellenic evidence either documentary or
archaeological. By way of preface, note only that Mr. Fotheringham's argu-
ment, being confined (with one exception l ) to Eusebian data, cannot lead to
any conclusion as to the veracity of Eusebius, but only as to his consistency ;
whereas my own object has been throughout to test by nou-Eusebian evidence
the respective veracity of the Excerpt and of the Canones..

Mr. Fotheringham's instances refer only to four thalassocracies (besides
Egypt) in or above the damaged part of the List — viz. to those of Caria,
Miletus, Phoenicia, and Thrace.

(«) It is difficult to see how the Carian mercenaries in the service of
Psammetichus, who did not begin to reign till 664 B.C., could be connected,
as Mr. Fotheringham suggests, with a Carian seapower which began in
735 B.C. and was over by 674 B.C., unless we are to assume that we have
here also one of those ingenious derangements of the entries by at least
ten years.

(b) In the case of Miletus, the more Milesian events Eusebius puts into
the neighbourhood of his Milesian seapower (748-730 B.C.) the worse for his
reputation as a historian. His date for Naukratis (748 B.C.) is as utterly out
of accord with fifth-century tradition 2 as it is with the archaeological evi-
dence as to the earliest occupation of Naukratis, and with the other
Egyptian evidence as to Bocchoris and the improbability of an Egyptian
seapower earlier than 664 B.C. Similarly, a chronologer who, in face of
Hdt. i. 74, was capable of putting Thales into the year 747 B.C., was capable
of any imaginable muddle. Except from these late chronologers, we do not
know much about the foundation-dates of Milesian colonies like Trapezus (or
was it Cyzicus, as Jerome says ?), which the Canones assign to 756 B.C. ; and
we know even less about the way in which the chronologers fixed these
dates. Mr. Fotheringham seems to incline 3 to the view that the Thalassocracy-

1 The ' Tyrian ' date for Cartilage, quoted cance of red ink (p. 81) with his treatment of
from Josephus, e. yip. i. 17, 18. the Pelasgic thalassocracy (p. S3), which seems

Hdt. ii. 178. to proceed on the opposite hypothesis.

Bui compare his discussion of the signifi-

'LIST OF THALASSOCRACIES ' IN EUSEBIUS : A REPLY. 129

list formed part of the framework of the Canoncs ; and if so, it would follow
that the other dates were accommodated to it. But if so, these other dates are
useless to determine the Eusebian view as to the proper place for the Milesian
seapower — i.e. to show the extent of the interval, if any, which separates
it from that of Phocaea below the gap in the list.

One point more, in regard to the foundation of Trapezus. If, as seems
admitted, the foundation of Phocaean Massilia in COO B.C. falls outside the
period of Phocaean seapower, what becomes of Mr. Fotheringham's
argument as to the connexion of Trapezus, if founded in 755 B.C., with a
seapower of Miletus which did not begin till 748 B.C.? It is the case of the
Carians of Psammetichus over again.

(c) The same criticism applies to the references to Phoenician
seapower. To argue from a date for Solomon's Hiram to a date for Dido's
Pygmalion is surely to confuse history and mythology. In any case, if I
understand Mr. Fotheringham correctly, the Eusebian date for the Phoenician
seapower has been adjusted to Josephus' 'Tynan' date for Pygmalion, on the
hypothesis that it was in the days of Pygmalion that Carthage was founded. 4
But who started this hypothesis ? There is nothing, so far as I know, to
support it in any Greek author before Alexandrian times, or in any extant
non-Hellenic author, and it disagrees by something like two centuries with
everything that is known, from the examination of Punic sites in Africa,
Sardinia, or Sicily, as to the upward date of Punic adventure in those lands.
The date, on the other hand, which I have proposed, on the clue furnished by
the List, by the limiting dates for Egypt, and by the Assyrian record, fits all
this archaeological evidence without difficulty, and disagrees merely with
post- Alexandrian chronographers.

(d) The Thracian conquest of Bebrycia illustrates, once more, the
uselessness of a chronological enquiry which does not go outside the
chronologer's materials. Was the conquest of Bebrycia put down under
972 B.C., because this date fell within the Eusebian limits for Thracian sea-
power, or were the limits of Thracian seapower adjusted to include the
Eusebian date for the conquest of Bebrycia? To argue from Eusebian data
alone is either to reach no conclusion or to argue in a circle.

To his mention of Trapezus already noted, Mr. Fotheringham adds that
' with the accuracy ' of this chronological system ' we are not concerned.'
But it is precisely its accuracy, which, from the standpoint of my article,
we are discussing. My whole contention is, in fact, that we know enough. 1 -\-
this time, from Egyptian and other non-Hellenistic sources, to be able to
assert that neither Eusebius, nor Trogus, nor any other Hellenistic or
Graeco-Roman chronologer knew anything of value about such matters as the
foundation of Trapezus, excepl in so far as he used at least a fifth- century
source: that between the filth century and Alexandrian times a thorough

4 Note meanwhile that we have been dealing (Jerome 1015 b.c. Ann.) and B{ rome :

here with only one cut of a number of dates 850 B.c. Arm.): and that the date for whicb

tV>r tlir foundation of Carthage, ranging from Mr. Fotheringham cites Josephus doi

1042 b.i . (Jerome : 1038 b.i . Aim. | to 1018 b.o. to appear in the I ' it all.

H.S. — VOL. XXVII. K

130 'LIST OF THALASSOCRACIES ' IN EUSEBIUS : A REPLY

obscuration of tradition took place ; and that it is only by going back either
to fifth-century historians, or to quite non-Hellenic data, that we can hope
to re-construct the early history of Greece. Whether the Excerpt from
Diodorus seems, or not, to preserve an echo of this earlier Hellenic tradition
is consequently a matter on which Eusebian evidence proves inevitably
nothing; especially if it be proved, as Mr. Fotheringham contends, to be
based itself upon that Excerpt.

If I were to attempt, in fact, to fix my position, on the whole question, in
a phrase, in face of Mr. Fotheringham's criticisms, I should do so best,
I think, by a re-arrangement of his own peroration. ' Inquiries into its
value,' I should say of the Eusebian List, ' are bound to be fruitless, unless the
restoration of the list is regarded as a matter of historical explanation,,
independent of the problem of textual criticism.'

John L. Myres.

NOTICES OF BOOKS.

Homer and His Age. By Andrew Lang. Illustrated. Pp. xii+336. Loudon :
Longmans, 190C. 12s. 6d. net.

Mr. Lang here returns to contentions advanced some years ago in Homer and the Epic.
His stimulus appears to have been supplied by the appearance of Mr. Leafs Iliad in 190? ;
and now he once more goes full tilt at the Separatists, asserting that these are much more
discrepant and inconsistent in statement than the Homeric lays themselves, and criticising
them in the light of more or less recent archaeology and )f a 'literary judgment,' which
is fortified by a wide knowledge of early epic literature in other languages than Greek.
Briefly he sets out to show that the discrepancies in Homer are not greater than would be
made by any one author in an age of vigorous and moving culture ; that the efforts made
to convict the poet or poets of archaism fail ; that the arguments used to prove archaism
in some passages or connexions and realism in others are absurd ; and that the lays do
represent very fairly a single civilisation. He sees no difficulty, in view of recent Cretan
discoveries, about supposing the poems to have been written down at an early age — an age,
in fact, much earlier than the Cyclic poets, not to mention Peisistratus ; and by the way he
makes much of Mr. Leaf's change of view concerning the relative probability of a Homeric
School and of a Peisistratean Recension. Judging by the ideas embodied in the Cyclic
poems, Mr. Lang would put the stereotyping of the Iliad (with which he is almost
exclusively concerned) some way back behind 800 B.C. He makes a good defence for unity
both in authorship and time ; but he seems to have overlooked, in advancing arguments
from the Cyclic poems, and comparing other Epics, one important fact, viz. that culture in
Ionia (where there is much reason to place the origin of the poems) had not necessarily the
same history as in Greece ; and that poems might have arisen at the same time on the two
sides of the Aegean, reflecting incongruous, but, in neither case, anachronistic ideas.
What would have been archaistic iu Greece in the seventh century was not necessarily
archaistic in Ionia. While regarding the Greek lands as endowed with too uniform a
civilisation in the post-Mycenaean Age, Mr. Lang also seems to treat the ' Mycenaean '
remains too much as one, and not to take sufficient account of possibly wide intervals in date
between, e.g. the later Palace at Cnossus, the Enkomi Treasure, and the ' Treasure from
one of the Greek Islands ' in the British Museum. When an author has to base his argu-
ments on the multifarious and often provisional statements of archaeologists groping their
way towards the light in the dim ages before history, and does not know the Rod',, a
himself, his foothold is often perilous in the extreme.

Pour mieux connaitre Homere. Par Michel Breal. Pp. viiL+309. Paris:

Hachette et Cie. [1907.] 3 f. 50.

M. Breal brings to the study of the Homeric question an acute mind trained in other
fields; and, as usual in such cases, his contribution i- fresh, original, and stimulating.

K -1

132 NOTICES OF BOOKS

He brushes aside the theory, which at one time found considerable favour in Germany,
that the Homeric poems, so to speak, grew of themselves, without any particular author : a
theory which only has to be stated in clear language to lose whatever plausibility it derived
from nebulous circumlocutions. Nevertheless he does not assign them to a single author.
His argument is that they are the product of a highly developed civilization, and were written
by a group of professional poets at some wealthy court in Asia Minor. Their supposed
simplicity of manners is conventional archaism ; their language is a mixed literary dialect,
which drew elements from various sources. M. Breal finds the necessary conditions
for such productions in the court of Lydia in the seventh century, and believes the poems
to have been written by a group of Greek poets, under Alyattes or Croesus, for recitation at
the creat games. This exposition of his views occupies only 130 short pages, and is not
worked out in detail ; but his arguments are quite sufficiently indicated, and are clearly
and attractively expressed. The rest of the volume is occupied by 144 short articles on
single Homeric words, on the lines of Buttmann's Lexilogus, but on a smaller scale.
It is a book to be recommended to Homeric students, and has the merit of being very
readable.

Anthologia Graeca Epigrammatum Palatina cum Planudea. Edidit H. Stadt-

mdeller. Vol. III. pars 1, Palatinae libri ix epp. 1-563, Planudeae 1. 1 continens.

Pp. vi + 584. Leipzig : Teubner, 1906.
Select Epigrams from the Greek Anthology. Edited with revised text,

translation, introduction, and notes, by J. W. Mackail. New edition. Pp. xi + 433.

London : Longmans, 1906. 14s. net.
The same. Greek text only. Pp. 175. 2s. net.

The satisfaction with which scholars will receive a new part of Stadtmiiller's critical edition
of the Anthology will be seriously damped by regret at the death of the editor before the
completion of his task. The present instalment, though of full size to rank as a volume
by itself, is described as the first part of vol. iii, and it appears before the second part of
vol. ii. It is needless to describe Stadtmiiller's work, which is indispensable to all serious
students of the Anthology. Unfortunately it appeared too late to be used by Prof. Mackail,
59 of whose selected 500 epigrams come from the portion of book ix included in the
present volume. Textual details are, however, the least important part of Prof. Mackail's
work, the appearance of which in a new edition is very welcome, since the original edition
(1890) has long been out of print. Without forgetting Symonds' excellent and inspiriting
essay it may safely be said that Prof. Mackail's selection, with its prefatory essay and
notes, forms the best introduction to the Anthology on its literary side. The new edition
differs only slightly in contents from its predecessor (ten epigrams have been omitted and
twenty added), but it has been carefully revised throughout. The only drawback is its
price, which will compel many who would have read it with profit and interest to deny
themselves that pleasure. It is some compensation that the Greek text is now separately
issued in a cheap and attractive form, which will make a delightful pocket companion ;
but it is a pity that Prof. Mackail's introduction should not be made more generally
accessible.

Isocratis Opera Omnia: recensuit E. Drerup. Vol. I. Pp. cxcix + 196. Leipzig:
Teubner, 190(3. 14 m.

A critical edition of Isocrates has long been needed, and the appearance of the first volume
of Dr. Drerup's work shows that the want is in a fair way to be supplied. It contains
the text of the first thirteen orations, with testimonia and critical apparatus, and full
prolegomena. The latter include not only a description of the textual materials
but a discussion of the dates and authenticity of the orations comprised in this volume.

NOTICES OF BOOKS 133

Textually, Isocrates is remarkable on account of the existence of two papyrus manu-
scripts of considerable length, which carry us back to a stage in the tradition before
the two main families of vellum MSS. diverged. One of these, the Marseilles papyrus
of the In Nicoclem, comes into use in the present volume ; the other, the British Museum
papyrus of the De Pace, will be of service in a subsequent volume. Dr. Drerup can be
trusted to make full use of it 3 since he has examined the original at length ; and since his
results have been revised and extended by Mr. H. I. Bell (whose complete edition of
the papyrus appeared in the Journal of Philology last year) there need be no hesitation
in accepting them as trustworthy. The continuation of Dr. Drerup's edition will be
awaited with interest.

Paralipomena Sophoclea : supplementary notes on the text and interpretation
of Sophocles, by Lewis Campbell, M. A. Pp. xv + 287. London: Rivingtons, 1907.
5*. net.

In this volume Prof. Campbell puts on record his final opinion on a large number
of passages in Sophocles in which his original interpretation differed from that adopted
by Sir R. C. Jebb. In some cases he frankly adopts Jebb's view ; in others he shows
cause for maintaining his previous opinion ; occasionally he offers a fresh explanation
altogether. Admirable as Jebb's edition is, it stands to reason that it cannot be final in all
respects, and no one lias a better right than Campbell, whose life has been spent in
the study of Attic literature, to express dissent and put on record alternative views.
And the tone which he adopts towards his great rival (or let us rather say colleague) is in
all respects admirable. Since the book consists wholly of short notes on a great quantity
of passages, it is obviously impossible to discuss it in detail here ; but it will have
to be taken into consideration by future editors, and it comes opportunely, since
the Oxford Press will soon, we hope, be including Sophocles in its series of classical
authors.

Bacchilide : epinici, ditirambi, e frammenti, con introduzione, comento, e appendice
critica. Di A. Taccone. Pp. li + 218. Torino : Loescher, 1907. Lire 3.50.

English students do not in general require any further edition of Bacchylides than Jebb's ;
but to those who desire a smaller or a cheaper book, which nevertheless contains a full
commentary on the poems, Taccone's work may be recommended as serviceable and
convenient. He has made full use of the previous literature on the subject, so that the
reader is placed in possession of the views that other scholars have taken of the restoration
or explanation of doubtful passages. The editor's original contributions are not large, but
he has carefully considered the work of others, and his edition should be very useful to
Italian students.

Adonis, Attis, Osiris : Studies in the History of Oriental Religion. By J. G. Frazer.
Pp. 356. London : Macmillan, 1906. 10*. net.

In the preface to this hook Dr. Frazer explains that the studies of Oriental cults are an
expansion of the corresponding sections in the Golden Bough, and will form part of the
third edition of that work. A careful student of the author's previous books will perhaps
be prepared for the general conclusions ; but the expansion is so considerable that readers
of the Golden Boiujh will find it necessary to revise their ideas of Adonis and other kindred
deities in the light of these later studies. Not only is much of the material new. but there
is also some difference in the treatment: in method, Dr. Frazer lavs greater stress on the
effect produced on Oriental religion by the natural features of the East ; in tone, there is

134 NOTICES OF BOOKS

a distinct change from the first (and, to a less extent, from the second) edition of the Golden
Bough. Dr. Frazer is no longer content to allow the many analogies between ancient cults
and modern Christianity to speak for themselves, and he frankly states his own conclusions
on some of the fundamental doctrines of the Christian religion.

The general argument of the book may be briefly stated in the author's own words
' under the names of Osiris, Tammuz, Adonis, and Attis, the peoples of Egypt and western
Asia represented the yearly decay and revival of life, especially of vegetable life, which
they personified as a god who annually died and rose again from the deid. In name and
detail the rites varied from place to place: in substance they were the same' (p. 5).
It is impossible, however, to do justice in a short notice to the learning and wealth of
illustration which support the argument. An interesting feature is the explanation of
customs or myths concerned with the burning of gods or kings, as possibly due to ' a
conception of the purifying virtue of fire, which, by destroying the corruptible and perish-
able elements of man, was supposed to fit him for union with the imperishable and divine '
(p. 100). Dr. Frazer suggests that, as men might attain to divinity by burning, so the
gods themselves might be refreshed and renovated by the ordeal of fire. The account of
the myth and ritual of Attis contains much that is new and striking, and the repulsive
part of the cult gains a fresh significance from Dr. Frazer's exhaustive treatment. His
conclusion, that the spread of Oriental religions in the West was one of the chief causes
that undermined ancient civilization is no doubt true ; but his attack on the ' selfish and
immoral doctrine ' of ' the commune of the soul with God and its eternal salvation as the
only object worth living for' (p. 194) will seem to many a prejudiced and unfair present-
ment of the ideals of early and medieval Christianity, which Dr. Frazer has in mind
(p. 195). The final chapters, on Osiris, are a very valuable contribution to our knowledge
of a subject of great complexity.

The Cults of the Greek States. By L. R. Farnell. Volumes III., IV. Vol. III. :
pp. xii + 392, with 35 Plates; Vol. IV. : pp. viii+454, with 51 Plates. Oxford:
Clarendon Press, 1907. 32s. net.

After an interval of more than ten years Dr. Farnell has published a second instalment
of his ' Cults,' the new volumes dealing with the Earth-goddesses (Ge, Demeter, Kore-
Persephone), Poseidon, and Apollo. The work has grown beyond the bounds which the
author originally contemplated ; three volumes in all were proposed in 1895, whereas in
the preface to vol. iii. it is announced that the book will be completed in five volumes, the
last to discuss Hermes, Dionysus, and minor cults. In vol. i. there was room for Zeus,
Hera, and Athena, while Demeter and Kore have practically filled an entire volume.
Except in the matter of expansion (and this is a distinct gain), Dr. Farnell has preserved
all the features which marked the two earlier volumes, the chapter devoted to the cults
being followed by chapters on the monuments and ideal type of each deity. It is
noticeable that Dr. Farnell now lays somewhat greater stress on the results of anthro-
pological study : quotations from Mannhardt, Lang, and Frazer are certainly more
numerous than before, especially in the treatment of Demeter and Kore, where, indeed,
the evidence of the Comparative method cannot be neglected. He does not, however, over-
estimate the importance of anthropology, and reminds us that 'its application to the higher
facts of our religious history might be combined with more caution and more special
knowledge than has always heen shown hitherto' (pre/, p. iv). The author himself is
eminently cautious in his own treatment of Demeter and of problems connected with the
Thesmophoi ia and Eletusinian mysteries. He will have nothing to do with a corn-totem,
remarking that there is no evidence for its existence in Greece (iii. p. 137) ; he does not
believe that the Thesmophoria can be explained by the theory that the invention of
agriculture and the cultivation of cereals were due to women ; and he is equally sceptical
with regard to the matriarchal hypothesis, by which Miss Harrison and others account for

NOTICES OF BOOKS 135

the Thesmophoria, among many other rites in Greek religion. His own view is that ' the
psychological explanation is more probable than the sociological' (p. Ill) ; women were
in charge of the Thesmophoria because they are apt to be more ecstatic and orgiastic, and
so hold a stronger magic, whereby they are more in sympathy with the earth-goddess,
Avhose generative powers resemble their own. His criticism of the matriarchate question is
a useful corrective to a theory which, as applied to Greek religion, has lately shown a
tendency to run riot. Equally sane is the discussion of the Eleusinia, a problem
exhaustively treated within the limits which the author imposes upon himself. On the
question, is there a secret worth discovery, and, if so, can it be discovered I Dr. Farnell
•comes to the conclusion that the fast, the mystic food, the passion-play, and the objects
revealed to the mystics produced, not a sense of absolute union with the divine nature, but
at least a feeling of intimacy and friendship with deities who were powerful in the nether
world, and could there reward their worshippers. The o-raxvs rede pia fj.evns, on which so
much stress has been laid, was not an object of worship, but one of the number of things
reverentially displayed.

The greater part of vol. iv. is assigned to Apollo, whose origin and cults are discussed
with good judgment. The title AuVios appears to belong to the oldest stratum of Apolline
religion, and as it can only mean 'wolf-god,' it implies a reverence for the wolf in a
hunting or pastoral stage of society ; but here again Dr. Farnell points out that this
respect need not be totemistic (p. 116). On the original significance of Apollo he is very
guarded, holding that the orthodox theory of an Aryan sun-god is a prion very possible,
but cannot be proved : the ordinary Greek, until the time of Euripides and Plato, did not
identify or associate Apollo with Helios, and Greek cult gives little support to the solar
theory. Dr. Farnell acknowledges that the reaction against the extravagances of the solar
myth may be pushed too far ; and it is perhaps a question whether his own treatment of
Apollo is not a case in point. The appellative Qolfios is certainly connected with <pdos y and
is most naturally explained as the epithet or name of a light-god. When the author
remarks that no one would maintain that the Sanskrit 'Devas' are all sun-gods, he seems
to miss a distinction : if, in one mythology, all gods are called ' bright ones,' the title
may well mean ' heavenly ' — a common appellative of a class ; but if, in another mythology.
a similar name is the exclusive property of a single god, it is hard to resist the inference
that this god was the 'bright one' par excellence, i.e. the sun. Again, Dr. Farnell
depreciates the evidence afforded by the name of &oi$r), of whom he says that 'nothing in
her legend or genealogy clearly reveals any solar trait.' There is of course no question
about solar traits ; but if Phoebe is a lunar goddess, as is implied by her later. identification
with Selene, we have thus indirect evidence that in primitive times the sun and moon
were called <t>oi(3o«r and <bol(iri respectively. It is however a fact, as Robert has already
pointed out (Preller-Robert, Grieeh. Myth. i. p. 231), that the sun-theory rests on no
evidence from cult, popular poetry, or art ; and it is well for us to be reminded that the
identification of Apollo with the sun really belongs to the same kind of ancient speculation
that derived Hera from the air. In any case, the true importance of Apollo lies in his
greatness as a political, social and ethical divinity, and in these respects Dr. Farnell s
admirable account leaves nothing to be desired. The ' ideal types' of Apollo are I
and discussed with discrimination, and show that the author is as sound in art-criticism as
in dealing with Greek religion. Scholars will welcome the completion of Dr. Farnell s
task, which is promised for nexl year; meanwhile we may congratulate him on a work
which cannot fail to be ranked among the most important of contributions to our
knowledge of Greek religion.

The Art of the Greeks. By H. B. Walters. Pp. 277 ; 112 plate-, and 18 figs, in
text. London : Methuen and Co. [1906]. 12*. 6d. net.

The author makes a general survey of the leading branches of Greek art. After two
chapters devoted to preliminary considerations, he discusses in turn the Architecture,

136 NOTICES OF BOOKS

Sculpture, Painting, Vases, Terracottas, Engraved Gems, Coins, and Metal work of the
Greeks. In each case, the elementary facts are stated, as far as may be, in chronological
sequence.

The illustrations are for the most part half-tone blocks, printed as plates. In many
cases, the results are brilliant.

Studies in the History and Art of the Eastern Provinces of the Roman
Empire. Written for the Quatercentenary of the University of Aberdeen by seven
of its Graduates. Edited by W. M. Ramsay. [Aberdeen University Studies, No. 20.]
Pp. xvi + 391. 11 Plates, 3 Maps, and numerous Illustrations in the Text. Aberdeen
University Press, 1906.

The title sufficiently explains the motive for the appearance of this book, although it is
somewhat misleading as an indication of the contents, seeing that they are confined to Asia
Minor, and indeed for the most part to Sir William Ramsay's special preserve, Phrygia and
the surrounding districts, Pisidia, Lycaonia, and Isauria. The volume is a remarkable
monument to the editor's genius for exploration and for inspiring others with his own
enthusiasm for a subject from which interesting results can only be extracted by the exercise
of much painful research. Of the seven contributors, the names of Miss Margaret Ramsay,
Mr. Callander, and Mr. J. G. C. Anderson, as well as of the editor, are familiar to readers
of this Journal. Miss Ramsay's contribution on Isaurian and East-Phrygian Art in the
Third and Fourth Centuries after Christ is, in fact, a development of her recent article on
that subject. It is the only one of the articles of much interest to the student of ancient
art. It is a conscientious and useful piece of work, but it is doubtful whether its value is
not diminished by the exaggeration of the importance of this local art. In the conclusion
that ' the mountain land of Northern Isauria . . . was the place of origin of a new kind
of decorative art, which spread widely over the Roman world ' we recognize an alarming
development of Dr. Strzygowski's theories. Mr. J. G. C. Anderson has an important paper
on Paganism and Christianity in N. Phrygia. The editor contributes a Report on
Exploration in Phrygia and Lycaonia, and also prints his Rede lecture on the War of
Moslem and Christian for the possession of Asia Minor. His most valuable contribution
is on the Tekmoreian Guest-Friends : perhaps an Anti-Christian Society on the Imperial
Estates at Pisidian Antioch. Some brief contributions in verse (English, Latin, and
Greek) give to this University publication a characteristically British touch. Though
some of the facsimiles of inscriptions leave much to be desired, the printing and
illustrations are good, and the volume as a whole thoroughly worthy of the occasion.

The Roman Forts on the Bar Hill, Dumbartonshire. By George Macdonald
and Alexander Park. With a note on the architectural details by Thomas Ross.
Pp. xii + 150. With 4 plates and 54 illustrations. Glasgow: Maclehose, 1906.
5s. net.

This is a clear and ably written account of one of the most interesting contributions that
recent excavations have made to our knowledge of Roman Britain. Considerable praise is
due to Mr. Whitelaw, the owner of the property on which the camps are situated, who
carried out tin; excavations at his own expense ; and he is also to be congratulated on
having entrusted the Beport to Messrs. Macdonald and Park. He must already have felt
amply rewarded when, on the very first morning, the workmen struck a well which had been
tilled up with a most extraordinary collection of antiquities. No such exciting find was
made in the subsequent excavation, but the plans of the Antonine fort and of Agricola's
amp which preceded it were recovered in some detail. Two inscriptions (one being a

NOTICES OF BOOKS 137

dedication to the Emperor Pius, and both mentioning the first Baetasian cohort) and some
curious, apparently prophy lactic, busts of freestone, not to speak of architectural fragments,
are among the stone remains. The most remarkable of all the relics is an admirably made
chariot-wheel of eleven spokes, found intact in a refuse-hole. The authors give good reason
for supposing that this and similarly made wheels are not Roman but native in origin.
Among the other remains are a fine set of leather shoes of various kinds ; a small series of
denarii (mostly shams, made of tin, apparently for the purpose of dedication to the gods !) ;
and some remarkable instruments of deer-horn, similar to others which were found to the
number of thirty-two in the armoury of Carnuntum. So far no one has succeeded in
explaining the object of these instruments, and the problem may be commended to the
consideration of readers of a mechanical turn of mind. The Report is well illustrated,
and more interesting reading than such reports usually are.

Antike Denkmaeler in Bulgarien. Edited by E. Kalinka (with nine colla-
borators). Vienna Academy, 1906. (Schriften der Balkankommission, Antiquar.
Abt. IV.) 220 pages, 162 figs, and many facsimiles.

This volume is a provisional Corpus of the architectonic, epigraphic, and sculptural
remains of Bulgaria, classified according to subject matter. 471 objects are admirably
catalogued, with adequate illustrations of all the sculptures; with facsimiles or illustrations
of all the inscriptions and with-eopious indices.

With the exception of a single item (No. 333, the early Greek stele of Anaxandros),
the whole of the works described appear to be of a late and provincial class. They are
now made conveniently accessible, and it is possible to take a general view of the extant
antiquities of Bulgaria.

Geschichte der Meder und Perser bis zur makedonischen Eroberung. [Handbiicher
der alten Geschichte.] Von Justin V. Prasee. Erster Band. Geschichte der Meder
unci des Reichs der Lander. Gotha, 1906.

This, the first volume of a History of the Medes and Persians, is divided into two parts.
The first deals with the Kingdom founded by the Medes. Here the chief points of
interest to the student of Greek history are the account of the Cimmerian invasion and
the discussion of the relative value of the authority of Hecataeus, Herodotus, and Ctesias
of Cnidus, the last named the physician long resident at the court of Artaxerxes. The
second part is devoted to the rise of the Persians and the foundation of their Empire under
Cyrus. The fall of Croesus and the expedition of Cambyses to Egypt are the events
which touch most closely upon Greek history- The volume closes with the revolt of the
false Bardes and the death of Cambyses. The book is valuable as presenting from the
standpoint of the Orientalist many events which are generally regarded solely from the
Greek point of view. The second volume will carry the history of Persia down to the fall
of the Empire of the Achaemenidae before the victorious Alexander.

Hat Dorpfeld die Enneakrunos-episode bei Pausanias tatsachlich gelost,
oder auf welchem Wege kann diese gelost werden? Einige Bemerkungen
zu Judeich's Topographie von Athen.Von Alexander Malinin. 35 pp. Wien. 1906.

This pamphlet on a threadbare theme justifies its existence by a new suggestion.
M. Malinin points out that in Pans. 1. 14. 5 the temple of Eucleia is described as dvddrjfia
kcu tovto dno Mr)8a>v, though no reference to the Persians occurs immediately before. In
fact the nearest reference to the Persians is in I. 8. 6, the passage about the statues of the

138 NOTICES OF BOOKS

Tyrannicides. Hence he infers that the chapters between have been interpolated ; and it
so happens that those chapters just contain the whole of the ' Enneakrunos-episode ' : if
they be omitted, the description of the agora is continuous. Thus Dr. Malinin claims to
have found indications of the wrong insertion of the 'episode' in this place. If he would
go on, and point out where is the place in which it ought to be inserted, he would have a
•complete and consistent theory as good as any that has been suggested on the matter.
Naturally those who see no need for assuming the insertion of the 'episode' will not be
convinced by his explanation.

Urkunden dramatischer Auffuhrungen in Athen ; mit einem Beitrage von
G. Ivaibel ; herausg. von A. Wilhelm. Pp. 279. 68 Abbildungen. [Sonderschr.
des Oest. Archaol. Institutes, Bd. VI.] Wien : Holder, 1906.

This volume, as its title and the names of its authors alone suffice to indicate, is of capital
importance and indispensable to all students of the history of Greek drama. The epigraphic
documents dealt with fall into three classes : (1) the list of victors at the Dionysia (I.G. ii.
971) ; (2) the Didaskaliai (I.G. ii. 972-976 and 1315 ; (3) the lists of victors, — tragic and
comic poets and actors — (I.G. ii. 977). The first is the famous list which in the first few
lines mentions Pericles and the poets Magnes and Aeschylus, recording the hitter's victory
with the Persae in 473-2 B.C. Dr. Wilhelm leaves the headline in the form ttp<ot]oi> km/jloi
rjcrau rw[i liopvcrax., rejecting with good reason Koehler's restoration tu>[v Tpaycoibwv xal to>v
Kw/xwiScoi', but not deciding between the many other possibilities. Three new fragments
are added to this list. The mention of Menander in tlie list of victors (ii. 977) enables Dr.
Wilhelm to give an interesting note on the date of his first appearance (he first repre-
sented 322-1, was first successful with the 'Opyfj, in 316-5) and on the date of the 'Euvtov
Ti/xcupoi'/xfj/or, showing that the king mentioned in Terence Heaut. 117 is not necessarily
Alexander the Great. Kaibel's contribution to the volume is an interesting discussion of
the chronological problems involved in the lists of victors. A series of appendices
deals with various side issues, and with recent works by Capps and Foucart bearing on the
history of the drama. The volume is characterised by the thoroughness and acumen which
is to be found in all Dr. Wilhelm's epigraphic publications, and is admirably printed and
indexed.

Histoire Sommaire des Etudes d'Bpigraphie grecque. Par S. Chabert.
Pp. 166. Paris : Leroux. 1906.

M. ('liabet's account of the study of Greek inscriptions from the earliest times (he begins
with Hellanicus) to the present day is very readable, giving somewhat more than the dry
bones of what does not at first sight seem a very attractive subject. The origin and methods
of the various attempts at a Corpus are fully explained, and the treatment generally is
sympathetic. Perhaps the writer is a little unkind to Fourmont, who, as Wilhelm has
recently shown, is not in all respects so black as he has been painted. Some minor inac-
curacies, especially in foreign names, are noticeable. Of books about or bearing on inscrip-
tions which are mentioned not at all or not in their natural place, we may note Bechtel's
monograph on Ionic inscriptions, Kern's Magnesia, Liiwy's inscriptions of sculptors, H.
-J. Bose's Inacriptiohes Graecae vetustissimae, the. third edition of Meisterhans by Schwyzer,
von Scala's Treaties, and Kirchner's Prosopographia. (Just's History of the Society of
Dilettanti is later than Michaelis' account, and the British School at Athens has a Govern-
ment grant.

NOTICES OF BOOKS 139

The Syntax of the Boeotian Dialect Inscriptions. By Edith F. Claflin.
(Bryn Mawr Diss.) Pp. 95. Baltimore, Lord Baltimore Press, 1905.

Miss Claflin's study of the Boeotian Inscriptions from the point of view of syntax is a
very careful piece of work, the real value of which will perhaps be more apparent when she
•or other scholars may have with equal patience similarly analysed the inscriptions of other
dialects. It will then he possible to obtain a true perspective of the syntactical peculiarities
of the various dialects. The present analysis reveals comparatively little that is peculiar to
Boeotian ; partly, it is true, because the brevity or official character of most inscriptions
hardly admits of much syntactical elaboration. In § 9. 4 it is a little surprising that the
writer should hesitate to decide between vocative and genitive in the grave-inscription
I.G.J. 149: KaXXm Myl{6)6our rv S' tv 7rp«o-[>' w] mipoSara. The genitive seems to us
undoubtedly the right interpretation. The reason for the use of the genitive in dating by
months might have been more clearly stated in § 9. 29 : a phrase like evvap X o> (ipxovros
.fxuvos Qeihweico of course expresses an extended period, not a point of time. It is
interesting to note that while eTrtWe is common in artists' signatures, the imperfect does
not occur.

Corolla Numismatica. Numismatic Essays in Honour of Barclay V. Head. Pp. xvi
+ 386. Oxford University Press, 1906. With frontispiece, 18 collotype plates, and
39 illustrations in the text. 30s.
That this sumptuous volume is of the nature of a Festschrift is sufficiently clear from its
title. The recognition it implies has been richly merited ; and it is pleasant to find that
the 'little crown ' is in every way worthy of the purpose it was meant to serve. Mr. G. F.
Hill has acted as editor, while the individual essayists number thirty in all. The continental
countries best represented are Germany and France, but tributes come also from Denmark,
Greece, Italy, and Switzerland. The variety of subjects handled is naturally very great.
There is hardly a side of ancient numismatics that is not touched upon, and consecpuently
no reader is likely to be sent empty away. In point of time, discussion ranges from the
Mycenaean age to the days of the Byzantine Emperors. Its geographical limits lie between
Gaul on thedie hand and India on the other. Esoteric disquisitions on problems of
metrology and technique are agreeably diversified by excursions into the less arid domains
of history and archaeology. As a rule, each of the essays has an importance of its own
in the present state of our knowledge, while a few of them are probably destined to be
permanently valuable. It is eminently fitting that so noteworthy a collection should be
associated with the name of Mr. Head.

In the circumstances it is not possible to give, within reasonable space, any detailed
indication of the character of the contents. Selection on the ground of merit might be
invidious. We shall, therefore, restrict ourselves to merely mentioning two or three of
the papers that may fairly be regarded as typical. None will attract more general interest
than M. Babelon's publication of anobol bearing the name of the Athenian tyrant Hippias.
Prof. Dressel's reconstruction of the temple of Matidia and the basilicas of Matidia and
Marciana is very tempting, although it depends upon the rehabilitation of a medallion
condemned as false by Eckhel. Colonel Allotte de la Fiiye's classification of the coins of
Persis marks a distinct advance towards the solution of a most obscure and difficult series
of questions. What M. Paul Perdrizet has to say of Nicopolis ad Mestum deserves the
particular attention of students of mythology. Mr. A. J. Evans opens up new vistas in
his account of Minoan weights and mediums of currency. Dr. Gaebler traces the history
of the famous coin collection of Queen Christina of Sweden. But we must forbear, adding
only that for the numismatic enthusiast there is scarcely a dull page in the honk. The
collotype plates are good ; and the portrait of Mr. Head, which forms the frontispiece, is
excellent. In the nature of things the compilation of an index was impracticable.

140 NOTICES OF BOOKS

Die antiken Munzen von Makedonia und Paionia. By Hugo Gaebler. Erste
Abteilung. Pp. viii + 196. With five collotype plates. Berlin: Reimer, 1906. 19 m.

It is just twenty years since Mommsen formulated his colossal scheme for the pub-
lication of a complete and articulated description of all known Greek coins. Work
was begun shortly afterwards under the supreme direction of Dr. Imhoof-Blumer. The
first half of Vol. I. appeared in 1898. Now, after an interval of eight years, we have
a second instalment of this monumental Corpus. It forms the opening section of Vol. III.,
the coinages dealt with being the national and provincial issues of Macedonia, including
Amphaxitis, Bottiaea, and Beroea. The book fully maintains the high standard set by
Prof. Pick's Dacien und Moesien. Dr. Gaebler is to be unreservedly congratulated. Not
only has he established a remarkable series of conclusions on a basis that promises to be
irrefragable ; he has set them before his readers with a lucidity and a precision that call
for the very warmest praise. His Introduction is peculiarly hard to summarize, for it
does not contain a single superfluous word. But its contents are so full of interest that a
bare resume must be attempted.

Philip II. had signalized the consolidation of his power by suppressing the various-
autonomous mints throughout Macedonia. Conversely, when Philip V. found himself on
the eve of a deadly struggle with Rome, he sought to enlist national feeling on his
side by sanctioning a revival of the national coinage. This national coinage came
to an end with the overthrow of the kingdom in 168 B.C. The victorious Romans divided
the country into four administrative districts or ' regions ' (jnepi'Se?), of which the 'first'
and the ' second ' struck both silver and bronze, while (so far as we know) the 'fourth'
struck bronze only and the 'third' never struck at all. The coinage of the 'regions'
began about 158 B.C., and the output of tetradrachms from the 'first' distiict must have
been enormous. In 150 B.C. occurred the revolt of Andriscus. The praetor, P. Juventius
Thalna, who was despatched against him, opened the campaign by seizing Amphipolis,.
where he struck tetradrachms with types borrowed from the preceding series, but showing
a marked difference in style, and bearing an olive branch (tfaXXo's) as a ' canting badge/
Presently Thalna was totally defeated by Andriscus, who in his turn proceeded to strike
tetradrachms at Amphipolis. The old types were still used ; but the obnoxious ' LEG[r/( ms
pro quaestore]' — which the Romans had introduced — was banished, and the head of
Artemis was bound with a laurel-wreath in token of victory. After the overthrow of
Andriscus by Q. Caecilius Metellus in 148 B.C., Macedonia became a Roman province.
Dr. Gaebler gives a careful list of the names of all the Roman magistrates who are
known to have been associated with the government of Macedonia from 148 B.C. down to
the reign of Philippus Senior, and we are thus provided with a convenient epitome of the
various changes that the form of administration underwent. During the republican era
the right of mintage was occasionally exercised by the Roman governors, as, for instance,
L. Fulcinnius and C. Publilius, quaestors of Metellus Macedonicus (148-146 B.C.).
Under L. Julius Caesar (93-92 B.C.) and C. Sentius Saturninus (92-88 B.C.) there was a
renewal of the silver-mining industry, and tetradrachms were minted very freely. The
bulk of this money bears the name of Aesillas, as quaestor. But there are two specimens
signed by his successor, Q. Bruttius Sura, as legatus pro quaestore.

Under the Empire the Macedonian coinage falls into two great classes, — imperial
pieces proper, and pieces without an imperial portrait. Dr. Gaebler has been aide to
accumulate for this period a mass of material that is practically exhaustive, and the
deductions he has been able to draw are correspondingly illuminating. Apart from their
direct bearing on Macedonian history, they have a wider interest in connexion with the
general questions that centre round the kolvov and the viu>Kopla, institutions that are more
familiar in Asia Minor than in Europe. The peculiarly 'agonistic' character of the later
coinage is well brought out, while an Appendix describes about thirty varieties of gold and
silver medals or medallions which also appear to have been connected with the dywves lepoi.
The huge gold medallions of the Tarsus find, for example, are to be associated with the
ayuiu lepos celebrated in honour of the veuxopia granted to Beroea in the reign of Elagabalus.

NOTICES OF BOOKS 141

It is matter for considerable regret that a discussion of the 'Aboukir' medallions is re-
legated to the Supplement. The Plates approach as near as may be to perfection.

Die Griechischen Mtinzen der Sammlung Warren. By Kurt Regling.
Pp. viii + 264. With 37 photographic plates. Berlin: Keimer, 1906. 40 m.

The 1769 Greek coins herein described belonged to Mr. E. P. Warren of Lewes. They
include 1016 that once formed the singularly choice cabinet of Canon Greenwell. It was
originally intended that the whole should pass into the Boston Museum of Fine Arts. But
this intention has only been partially fulfilled ; some of the coins remain in Lewes, others
were sold in London in 1905. Prior to the collection being thus broken up, it was decided
to compile a permanent record of its contents. Those responsible were fortunately able to
secure the services of Dr. Regling for the purpose, and the result is the handsome volume
now under notice, — a book that no student of Greek numismatics can afford to neglect.
The issues of the best period, from all parts of the Greek world, are fairly well represented.
But specimens of the electrum of Cyzicus and of the gold and silver of Sicily, Abdera,
Lampsacus, and Cyrene are specially numerous and important. For the most part the indi-
vidual coins have been selected from the point of view of one who combined the instincts
of the collector with the tastes of the scholar. As a result, we get many examples that are
very fine, a considerable number that are rare, and a few that are unique. Regling's de-
scriptions are characterized by the care and thoroughness that were to be expected from so
competent a numismatist. Particular value attaches to identifications of dies with those
of coins published elsewhere. Useful notes abound, and there are frequent references to
recent numismatic literature. Taken all over, the Plates are good.

Terina. Programm zum Winkehnannsfeste. By Kurt Reglixg. Pp. 81. With 3 Photo-
graphic plates and 2 cuts in the text. Berlin : Reimer, 1906.

The beautiful series of didrachms struck at Terina is a great favourite with all lovers
of Greek coins. Nearly twenty-five years ago (1883) many of them were illustrated in a
paper published in the Numismatic Chronicle by the late R. S. Poole. The present
monograph is far more complete than anything hitherto attempted. It is practically an
exhaustive list of all known specimens. The number of distinct varieties catalogued
(apart from plated coins) is 84. Guided by stylistic considerations, combined with
a minute study of the dies, Regling distributes these over seven periods covering the
years between 480 and 356 B c. The most interesting of the periods is that generally
associated with the handiwork of an engraver*. The opinions of other numisma'i-N
regarding this artist are passed in review, and at least one fresh piece of evidence is
adduced. The final conclusion is that it is an abbreviation of the name Phrygillos, and
that the little bird, which appears both at Terina and at Thurii, is a ' canting bad-.- '
(<pf>vyi\os). An examination of the types suggests that the well-known female figure is
neither a nymph nor a Siren, but Nike. — an explanation that has already had its advo-
cates : any peculiarities in the representation are to be accounted for by assuming a
' syncretism ' with the city goddess, Terina. Incidentally, nrca 300 B.c. is fixed upon as
the most probable date for the issue of the silver tetrobols. The bronze coinage
receives brief discussion. Altogether, this 'Programm' is an excellent bit of work. The
Plates are admirable.

142 NOTICES OF BOOKS

Grammatik der griechischen Papyri aus der Ptolemaerzeit : Laut- und
Wortlehre. Von E. Mayser. Pp. xiv + 538. Leipzig: Teubner, 1906.

So long ago as 189S Prof. Mayser published, in a Heilbronn Progrannn, the first part of a
grammar of the Ptolemaic papyri, dealing with the vowel phenomena presented by them.
Since that date the materials have greatly increased, mainly through the publication of the
Tebtunis papyri and the third part of the Petrie papyri (the Hibeh papyri were published
too late to be taken into account), and Prof. Mayser has now rehandled the whole subject
on a larger scale, — a scale so large, indeed, that it is not likely that any word in the
published Ptolemaic papyri has escaped his notice. The present volume deals with
phonetics and accidence ; a second is promised upon the syntax. In a department of
learning where the materials are constantly increasing so rapidly as is the case with Greek
papyri, it is impossible to expect finality ; but as a very full conspectus of the extant
phenomena Prof. Mayser's book will be a useful work of reference for some time to come

Life in Ancient Athens. [Handbooks of Archaeology and Antiquities.] By
T. G. Tucker. Pp. xiii + 212. With 85 illustrations and a map. London:
Macmillan and Co., 1907.

This is in many ways quite a remarkable little book. It is written in an extremely fresh
and attractive style, which is well maintained throughout. It is quite unpretentious ::
not a single reference to authorities is given in the body of the work. Yet there are
abundant evidences that the author has an intimate acquaintance with Greek literature-
and no slight knowledge of Greek archaeology. The period dealt with comprises the
fifth and fourth centuries B.C., and all the varied interests of Athenian life at that period
both public and private, are vividly presented. The different chapters describe the
buildings the Athenian saw around him, how he spent his time day by day, what he wore,
how he brought up his children, what was his religious belief, and other similar interests
of life. The main characteristics of Athenian art are briefly dealt with. Perhaps we
should like to have heard a little more about the average Athenian's attitude towards
Imperial questions. It may be noted also (a trivial point) that the spelling of place-names
in the map does not always agree with that adopted in the text. The illustrations are,,
with one or two exceptions, excellent.

The Princes of Achaia and the Chronicles of Morea. By Sir Rennell.
Rodd. 2 vols. 8vo. Pp.655. London: E. Arnold, 1907. 25.^. net.

Sir Rennell Rodd unfolds an interesting story of mediaeval Greece, which has probably
been known to English readers chiefly through the more limited studies of Finlay, Tozer,.
and Bury. Gibbon, who had not the necessary material, dismissed the subject with a
lordly sentence : — ' I shall not pursue the obscure and various dynasties that rose and fell oil
the continent or in the isles.' The author of these volumes has naturally made full use of
the invaluable researches of Carl Hopf, but he has made investigations of his own with
great enthusiasm and patience. He gives in an appendix an account of the mediaeval
Chronicle of Morea, a record truly prosaic as a poem yet indispensable, though at times
treacherous, as an authority. Prof. J. Schmidt's edition (London : Methuen, 1904) is
referred to, though it seems to have been published too late for the author's ready use.
(There is a still more recent critical study of the Chronicle by A. J. Adamantin, published,
at Athens, 1906, and noticed in Byz. Zeitxchrtft, 1907, p. 335.)

Sir Rennell deals with a period that extends from the Latin conquest of Constantinople-
in 1204 to the fall of the Eastern Empire in the middle of the fifteenth century. We have
the whole history of the princedom of Achaia — practically the Peloponnese — and glimpses

NOTICES OF BOOKS 14$

of the rulers of Epirus and Thessaly, of the Dukes of Athens, and of the rather distracting
governors of Euboea. The scene is often laid in such classic lands as Elis and Arcadia,
with a Greek and Slav population in the back-round. French knights like the Ville-
hardouins and the sinister Charles of Anjou, or his bailies, impose on alien peoples the law
of the West— not harshly, indeed, yet without enduring results. Points of feudal etiquette
are referred to the French King ; the Pope intervenes ; the Doge of Venice is ever on the
alert ; and the Eastern Emperor plays the part— sometimes a small one— for which he lias
been set down. The most interesting portions of the work, for the general reader, are those
describing the conquest by the Villehardouins. Sir Rennell writes very well, though his
narrative at times— perhaps inevitably— rather tends to dry chronicle and the narration
of family history. There is always however a useful statement of facts clearly set
forth. The book contains some genealogical tables and an interesting map of mediaeval
Greece, where the reader can find Clarenza and Clairmont and Andravida, and other places
famous in Prankish story. Perhaps the author may some day be ab'e to give us a
supplementary volume, or at any rate a portfolio of illustrations, with' letter press, con-
taining pictures of some of the old strongholds that form such a romantic feature in the
feudal scenery of Morea ; and a selection from the coins, and possibly some other
illustrative matter, would be welcome.

Bi.p\io0T|KT| Mapao-Xfj. MeXeVae nep\ tov ftiov Kai t?js yXoiaar]^ tov 'eWtjvikov Xnov, wro N. I\

noXirou.

napoi(xiai. A', pp. tt' + 600, 1899. B'. pp. 127 + 699, 1900. 1", pp. 686, 1901. A', pp. 686,
1902. 'Ev 'hdfivuK, tvttOk n. A. ?:aKe\\apiov. Price, each volume, 8 f.

These four volumes are a first instalment of a complete collection of modern Greek
and Byzantine proverbs, brought together by Professor Politis from all sources, published,
unpublished, and oral, and fully annotated. The most important word has been taken in
each proverb, and the collection arranged alphabetically under these headings. The scale
of the work is very large, and the last proverb in the fourth volume is entered under a
heading no further down the alphabet than e'Xe<S. General conclusions on the whole
subject are promised at the completion of the work. In the first volume, before beginning
his own collection, the author has printed seven unpublished collections of Byzantine
proverbs from MSS. at Munich, Corfu, Athens and Jerusalem, and a bibliography, which
includes foreign as well as Greek proverbs. The work is far from being a mere compilation
of material already published. Besides the proverbs he has collected personally, he has
used lists of proverbs sent to him for this purpose from all parts of the Greek-speaking
world. Of these the most important are 3386 from Lesbos, and 2371 from Cephalonia.
Each proverb is explained and illustrated by comparisons drawn from a wide field.
Owing to the lack of a dialect Dictionary of modern Greek, and the number of
unusual or local words used, these explanations are not the least valuable part of
the work.

The proverbs are recorded, as far as the sources allow, in the local dialects, and,
whilst they unavoidably, if only from the inadequacy of the Greek alphabet, fall short
of complete accuracy in this respect, none of the native colour has been removed by
any translation into the purified language. The sources used are so wide and the arrange-
ment so methodical that, as far as modern proverbs are concerned, it does not seem
possible that it can be superseded, and no fresh discoveries of Byzantine MS. collections
are likely to add much of importance. The removal of the National Library caused the
work to be broken off in 1?04. The interval has been employed on a collection of Uapa86(reis
or modern Greek traditions, and, when this is finished, the publication of the Proverbs
is to be resumed. The volumes are well printed in the same format as the rest of the

15 1 JXiofirJKT) MopncrXr;.

144 NOTICES OF BOOKS

Bip\io0T|KT] Maf>a<rXf]. MeXeVat ivep\ tov (Blov Kai. rfjs yXoxTcrrjs tov 'eWt^vikov haov, vno

N. T. IIoXitou.
IIapa8do-€is, Me'poy A', pp. 1-628 ; Mepos B', pp. 629-1348. 'Ev 'AOrjvais, rinrois II. A. 2a*eX-

\apiov, 1904. Price including M/po? A', 20 f.

These two volumes, of which the first contains 1013 legendary stories, and the second
notes on Nos. 1-644, will be followed by a third containing the rest of the notes and the
Prolegomena. The whole is to form a part of the MeXeW, which Professor Politis is
publishing in the Maraslis library. The appearance of these books, with their good printing
and moderate price, is due to the enlightened generosity of Mr. Gregorios Maraslis.

The traditional stories in this collection are both from published and from oral
sources. The latter are transcribed as they were told, in the genuine popular language
untouched by the written tradition, and more or less coloured by the peculiarities of the
local dialect. Those taken from printed sources, many of them foreign, have been recast
in the popular form, in which they might have been recounted by peasants using their
local dialects. These latter are marked with an asterisk. This distinction will no doubt
be mentioned in the forthcoming Prolegomena : at present it can only be made out by
noting the nature of the source as given in the notes. It is obviously of capital value for
anyone who would use the book as material for the study of the dialects.

The traditions are arranged under twenty-nine heads : — legends historical and local,
legends of gods, saints, and heroes, of the stars and elements, of plants, animals, and wild
beasts, of dragons and serpents, of treasures guarded by negroes, of ghosts and haunted
places, of uncanny creatures, Kalikantzaroi, Neraides and Lamias, of witches, of the devil
and apparitions, of sicknesses, of the fates, of corpses, vampires, death, and the underworld,
with a final section containing aetiological stories.

Such a collection cannot from the nature of the subject be complete, and anyone who
has enjoyed a part of the confidence of Greek peasants could add a few more items, but it
gives samples probably of every kind of legend. The author has done good service, not
only in printing the large number of legends he has himself collected, but in gathering
together the published stories, which were scattered over a great mass of literature, much
of it very inaccessible. The skill, with which he has retold these in popular form restores
to them much of their life, sadly lost in foreign books, or in the purified language of
modern Greek writers.

The legends themselves, as products of the popular Greek fancy, are of as much
interest as the Romaic folk-ballads, and sometimes deal with the same subjects. The
taking of Saint Sophia, the hero Digenes, the woman buried beneath the bridge, appear in
tradition and folksong alike. The mass of material, both in text and notes, bearing on
popular mythology makes the book indispensable for the student of Greek folklore, and
the charm of the stories will appeal to the general reader with some knowledge of modern
Greek. The notes are very full, and embody comparisons and illustrations drawn from a

wide field.

*****

The following hooks haoe also been received : —

Atkinson (T. D). A Glossary of Terms used in English Architecture. Pp. xxiv + 320;

265 cuts. London : Methuen, 1906.
Duff (J. W.). Homer and Beowulf. Pp. 25. 1906.
Ferrero (Gugl.). The Greatness and Decline of Rome. Transl. by A. E. Zimmern.

2 vols., pp. viii + 328, vi + 389. London : Heinemann, 1907. 17s. net.
Gilliard (Charles). Quelques Reformes de Solon. Pp. 324. Lausanne : Bridel, 1907.
Hoste (M. R). Nausicaa. Pp. xi + 56. London : Nutt, 1906.
Juvenal. Satires, ed. A. F. Cole. Pp. xii + 382. London : Dent, 1906.
Plato. Menexenus, ed. J. A. Shawyer. Pp. xxxi + 36. Oxford: Clarendon Press,

1906. 2.s.
Schmidt (M. C. P.). Kritik der Kritiken. Pp. 37. Leipzig : Diirr, 1906.
Sophocles. Antigone. Transl. by R. Whitelaw. Pp. 1 + 56. Oxford: Clarendon

Press, 1906.

J. H. S. Vol. XXVII (1907) Pl. XIII.

EARLY SELEUCID PORTRAITS.

J. H. S. Vol. XXVII (1907) Pl. XIV.

EARLY SELEUCID P< )RTRAITS.

EARLY SELEUCID PORTRAITS.

II.

[Plates XIII., XIV.]

In a recent number of the Journal 1 I had occasion to indicate some of
the difficulties surrounding the identification of the royal portraits that
occur on silver coins accompanied by the simple inscription BAZIAE2Z
ANTloXoY- The object of the article in question was to advocate a
change of tactics in dealing with the problem, — to urge the desirability
of concentrating attention on well-defined groups which should be sub-
jected to a close and comprehensive scrutiny. As an illustration of the line of
treatment proposed, there was selected for detailed examination the set of
coins composed of tetrabrachius on which the diadem worn by the kino- is
furnished Avith wings. While certain of the inferences tentatively suo-o-ested
on the strength of this examination have not been universally accepted, 2
the more positive and important of the conclusions reached remain uncon-
troverted. I would single out the following points as being now fairly well
established : — (1) A large class of coins previously assigned to Antiochus Hierax,
or alternatively to Antiochus III., really belongs to Antiochus II. (2) The
pieces of which it consists give us a portrait of Antiochus II. which Ave may
confidently adopt as a ' standard ' likeness, a criterion that in his case Avas not
previously available for purposes of classification. 3 (3) Most of the gold money
of Antiochus II. Avas struck in Central Asia at a mint or mints Avhich had also
been active under his father. (4) Whether Ave shall ever be able to recognize
the portrait of Hierax or no, the majority of the tetradrachms with the winged
head form part of his coinage, having been minted at Alexandria Troas.
(5) These tetradrachms, taken in conjunction with a number of others
of similar 'spread ' character, issued from the mints of Cyzicus, Lampsacus,
and Abydus, gives us a fair idea of the ' sphere of influence ' which Hierax
dominated before 229 B.C.

AYhere a first experiment has proved so fruitful in interest, it is perhaps
justifiable to embark upon a second. And there lies ready to hand
another group of coins sufficiently marked in character and sufficient lv
limited in extent to make investigation comparatively easy. I mean the
rare tetradrachms which have on the reverse a figure of Heracles resting.
The hero is seated with his lion's skin beneath him. while his right hand
grasps the upper end of his club, which stands upright in front of him.
This type constitutes a striking innovation in the coinage of the Seleucidae.
It is intrusive, in the sense that the familiar figure of Apollo on the

1 J. U.S. xxiii. pp. !<2 fl'. J. U.S. xxv. pp. 101 ri'., and H. von Fritze in

2 Discussion lias chiefly centred round the Berl. Phil. Wcchenschr. 1906, Xo. 27 (p. 731).
portrait' of Hierax; see A. J. 13. Wace in *-See Bunbury, Num. Chrdn. 1883, p. 7-.

H.S. — VOL. XXVII. L

Ii6

GEORGE MACDONALD

omphalos is speedily restored to its place of honour. That the variation was
a local as well as a passing phase of numismatic fashion was long ago
pointed out by Dr. Imhoof-Blumer, whose single remark upon the subject
is the truest and most valuable observation that has so far been made
about these tetradrachms : they were all struck in Ionia or in Aeolis. 4

Other numismatists besides Imhoof have had their attention turned to
our group. 5 In no case, however, have the materials for comparison and
study been anything like complete. As a consequence the results obtained
have been inconclusive or unreliable. Bunbury and Babelon, for instance,
were at pains to argue (as agaiust Gardner) that the type of the seated
Heracles was peculiar to Antiochus II. So far as the coins then known are
concerned, it is certain they were right. But the discovery of the piece
which stands first upon our list has given an entirely different aspect to the
question, and has at the same time finally disproved the view of Six, who
believed that the whole of the Heracles tetradrachms belonged to the rei^m
of Antiochus Hierax, and that they represented the coinage of ' Alexander,
brother of Laodice,' having been issued by him while he was holding Sardis
in the interests of his 'nephew.' The identity of the head upon the
following is so clear that we need not hesitate to attribute it to the first of the
Antiochi.

Antiochus I.

I. 7 Head of Antiochus I., r., dia-
demed ; border of dots.

BAZIAEQZ Heracles, naked, seated
ANTloXoY 1. on rock, his hair
bound with taenia ; underneath
him is his lion's skin, one end of
which is brought up so as partially
to cover his r. thigh ; his r. hand
grasps the handle of his club, which
stands upright in front of him,
while his 1. is placed behind him
on the rock ; beneath s A ; to 1.,
beyond inscr., one-handled vase
r., below which, one above another,
(51 and f^ .

PI. XIII., 5=B.M.= Greek Coins of a WeH-hnoum Amateur, Lot 230 (PL vii) = Rcgling,
Sammlung Warren, p. 202, Xo. 1297 (Taf. xxx.).

4 Mon nates grecques, p. 426. Cf. Bunbury,
Num. Chron. 1883, p. 78, footnote.

5 Gardner, Scleucid Kings of Syria, pp. xv. f ;
Bunbury, I.e. pp. 77 ff. ; Babelon, fioi&de Syric,
pp. lx. ff. ; Six, Num. Chron. 1S98, pp. 233 f.,
etc.

6 See J. U.S. xxiii. p. 116.

7 For convenience of reference the coins in the
particular group under examination are
numbered consecutively, irrespective of the
king whose portrait they may bear. All of them

are Euboic-Attic tetradrachms. Where different
specimens are catalogued under the same
number, it is to be understood that they are
from the same dies on both sides. Where the
mathematical sign of equality is employed, it
means not merely that the specimens thus con-
nected are from the same dies, but that they are
identical.

8 The use of the word ' beneath ' in a descrip-
tion implies that there is no exergual line.

EARLY SELEUCID PORTRAITS. 147

This remarkable tetradrachm, formerly in the Warren Collection, is now
in the British Museum. Mr. Talbot Ready, through whose agency it was put
upon the London market, informs me that it was found some seven years
ago in the Lebanon. Though it has been published twice previously, it has
not yet attracted the notice it deserves. As a glance at the Plate will show,
the russred features of Antiochus Soter are unmistakable. The coin was
therefore struck not later than 261 B.C. And it came from the mint of Cyme in
Aeolis. The one-handled vase on the reverse would of itself have been sufficient
to prove that this Avas so. It was the ' town-arms ' or Trapdcnj/xov of Cyme.
As such, it figures as a symbul on silver coins of various periods, notably on
tetradrachms with the types of Alexander the Great, 9 and on the familiar
'spread' tetradrachms of the second century B.C.; on Plate XIIL, 2, for
instance, it will be seen between the legs of the horse on the reverse. On bronze
coins it occurs not merely as a symbol but also as a type ; in such cases it
shares the usual fate of the irapda-qfiov and is relegated to the reverse, 10 as on
Plate XIIL, 1. Standing alone, then, the one-handled vase upon No. 1 would
have furnished ample evidence of origin. But there is a link that is even
closer and more interesting. The two monograms that are placed below it
obviously denote the names of magistrates, and one of them is very uncom-
mon. Now exactly the same combination appears on the reverse of an autono-
mous silver coin of Cyme, which is also, as it happens, in the British Museum.
The following is a description.

Head of the Amazon Cyme r. ; hair I KY (above) Bridled horse standing-
rolled and tied with riband ; | r., with 1. forefoot raised ; between
border of dots. its legs, fft; in front, p|.

PI. XIII., 6-B.M.C. Troas, Aeolis, and Lesbos, p. 109, No. 58 (PI. xx. 16).

This piece, which has a weight of 10-47 grammes, is one of a very rare
class. It was originally published in 1892 by Mr. Warwick Wroth, and was
regarded by him as a didrachm of the Persic standard, probably belonging to
the period from 250 to 190 B.C. 11 Curiously enough a second example, struck
from different dies and weighing only 9'36 grammes, was made known in the
same year by M. Babelon, who considered it to be a light Rhodian tridrachm,
minted between 258 and 202 B.C. 12 A third specimen, bearing different
monograms and weighing 10'55 grammes, had been described nine years earlier
by Dr. Imhoof-Blumer, 13 who, however, expressed no opinion as to its age ;
it was then in the cabinet of Sir Edward Bunbury, and is illustrated in the
Sale Catalogue of his collection. 14 It will be noted that there has beeu a
tendency to assign this autonomous group to the days uhen the hegemony

" Miiller, Nos. 943 IF. i*. 12.

10 Coin Types, pp. 123 II'. u Moimaus greeqws, p. 272.

ii Num. Chron, 1882, p. 17. ; s Portion, Lot 131 Plate II.'.

)- R c. Num. 1892, p. 116, No. 12, PI.

L 2

148 GEORGE MACDONALD

of Western Asia Minor had passed from the hands of the Seleucidae. The
connection now established with No. 1 of the Heracles series pushes it back
to the reign of Antiochus I. It is true that the identity of monograms may
be a coincidence ; in certain Hellenic cities the monetary magistracies would
seem to have been hereditary. 15 But I think a comparison of the two pieces
concerned — they are placed side by side upon Plate XIII. — will convince any
experienced eye that they are practically contemporaneous. Special signi-
ficance attaches to the presence, on the obverse of both, of the border of dots,
an adjunct that fell into disfavour in this part of the Hellenic world about the
middle of the third century B.C. 16 It may appear strange that regal and
autonomous coins, — of different standards too, — should have been issued from
the same mint in one and the same year. But, although we know too little of
the circumstances of the time to hazard an explanation, attention will be drawn
presently to what is possibly a parallel. Meanwhile we must content ourselves
with noting that the gap in the mintage of Cyme is less absolute than was
formerly supposed. 1 '

Apart from this fitful and uncertain gleam of light, the discovery of
No. 1 has an important bearing on the discussion of the remaining tetra-
drachms of the Heracles group. Although these have hitherto been assigned
by general consent to Antiochus II., they fell to him only as ultimas hacrcs.
'.It must be admitted,' says Bunbury, 'that the reasons for attributing this
particular group of coins to the second Antiochus, instead of his predecessor
or successor, are extremely slight.' 1S Noav we are for the first time in a
position to provide a solid basis for the attribution, and so to secure a fresh
set of well-authenticated portraits of the king. That the whole series is
homogeneous does not admit of doubt, and will become even more evident
a? we proceed. This being so, the identification of the head of Antiochus I.
gives us a fixed point. The younger head can only be that of his son, and
the probable period of issue is circa 201 B.C. The new king was at that
time twenty-four, an age that agrees perfectly with his appearance, not indeed
in all of the portraits, but certainly in those of them that w r e can, upon
other grounds, accept as being 'standard' likenesses. The nature and cause
of the variations hinted at will become apparent immediately.

Inclusive of No. 1, I have succeeded in bringing together twenty-five
different varieties of Heracles tetradrachms, several varieties being repre-
sented by more than one specimen. It is an agreeable duty to acknowledge
the kindness and courtesy of owners and custodians who have furnished me
with casts. And it is a further pleasure to add that the cost of supplying
adequate illustrations has been met by a grant from the Research Fund of
the Carnegie Trust for the Universities of Scotland. Some general remarks

15 See ./. TT. S. xxiii, p. 101. and 190 B.C. the city 'docs not seem to have

16 J. II S. xxiii. p. 115 — where, by the way, struck any money whatever.' {Hist. Num.
there is an obvious misprint of 'second' for p. 479.)

'third.'. 1S Num. Chron. 18.83, p. 77.

17 In 1887 Head remarked that between 500

EARLY SELEUCID PORTRAITS. 149

will facilitate the proper understanding of the detailed descriptions that are
to follow. To begin with, it will be found that the Heracles coins of
Antiochus II. group themselves naturally into three classes, each class
characterized by a distinctive mark or marks enabling us to assign it to a
particular mint. A peculiarity common to all three classes calls for very special
notice. At the head of each are ranged one or two pieces unexceptionable
in style and execution. Those that succeed them are simply more or less
degenerate copies. The result is an apparently wide variety of portraiture, the
real meaning of which only becomes intelligible when we have something
like a complete sequence before us. In every case the earliest coins were
produced by skilled engravers ; as the original dies wore out or broke, they
were replaced by imitations which betray the hand of inexperienced workmen,
but which were destined in their turn to serve as models for even lower depths
of deterioration. We cannot, of course, be certain in any instance that we
possess all the links in the chain. But the surviving evidence is quite
sufficient to demonstrate the broad truth of the statement just made. The
same phenomenon has been observed elsewhere, in connection with the
money of cities that have no continuous minting tradition stretching back to
fairly early times. Crete supplies quite a number of examples. 10 And it is
significant that the process described manifests itself very clearly in the
small group of autonomous coins of Cyme to which we had occasion to
allude a page or two back; 20 the British Museum piece is admirably
executed, that in the Bibliotheque Nationale is not nearly so satisfactory, the
Bunbury specimen is a long way behind both. It would seem as if cities
that took to minting at a comparatively late epoch were in the habit of
importing from well-established centres, not trained workmen and designers,
but pattern dies; these dies were used as long as might be; when they
ceased to be serviceable, recourse was had to the talent of local 'artists.'
However that may be, the phenomenon we have been discussing introduces a
fresh complication into the problem of Seleucid portraiture. It is obvious
that, where it occurs, the only likenesses we can regard as typical are those
that ojDen a series. One other remark is worth making before we enter on a
consideration of the individual varieties. The list will be found to illustrate
two of the technical points to which attention was drawn in my former
paper. 21 The border of dots on the obverse is, as we might expect from the
date of issue, universal ; and the obverse die had, as a rule, a much lonsrer
life than the reverse. Having thus cleared the ground, we may proceed to our
enumeration.

19 See Svoronos, Numismatiquc de la Crite See supra, p. 147.

ancienne, Atlas, passim. - l J.H.S. xxiii. pp. 99 f. and p. 103.

150

GEORGE MACDONALD

Antiochus II.

Class A. — Struck at Cyme.

Head of Antiochus II.,
denied ; border of dots.

Subdivision (a).
., dia-

Pl. XIII.

BAIIAEQI Heracles seated 1. on
ANTIoXoY rock as in No. 1; no
taenia; although there is no ex.
line, the ground is indicated run-
nine from the edge of the rock below
the hero's feet ; beneath, one-
handled vase 1., and p|; between
legs of Heracles, A ', to r. of lower
part of rock, jh.
7 = Greek Coins in the Hunter tan Collection, Vol. III. p. 19, No. 1.

3. Similar.

PI. XIII., 8 = Berlin.

Similar

rock.

Subdivision (J3).

but 4^ beneath r. end of

Head of
denied ;

Antiochus II.
border of dots.

dia-

Pl. xill., 9-B.M.C. p. 14, No. 8.
5. Similar.

BAIIAEQI Heracles seated 1. on
ANTIoXoY rock as on No. 1 ; he
wears taenia, and there is no at-
tempt to indicate the ground ; to 1.,
beyond inscr., one-handled vase 22 1.,
with traces of monogram below
(A '•) i to 1. of lower end of club,
^ ; to r. of lower part of rock, £J2.

BASIAEQE Similar type; to r. of
ANTloXo[Y] lower part of rock $Jj ;

no symbol and no other monograms

visible.

PI. XIII., \0 = Auctions-Catalog Hirsch, XIII. p. 277, No. 4439.

Similar ; style slightly less refined ;
dots in border larger.

PI. XIII.

BAIIAEQZ Similar type; style
ANTIoXoY slightly less refined;
to 1., beyond inscr., one-handled
vase 1. ; beneath, E ffl ; to 1. of lower
end of club, A ; to r. of lower part
of rock ^.
11 = Berlin (Irahoof, Mown, grccq., p. 236, No. 28) ; Berlin (Lobbecke).

22 Here (and in several other cases) my
description differs in some details from that

already published. All such corrections and
additions have been most carefully verified.

EARLY SELEUCID PORTRAITS

151

7. Similai

B A Z I A E Q Z Similar type ; to 1., be-
A N T I o X o Y yond inscr., one-handled
vase (?) ; beneath, ^ J no other
monograms visible.
PI. XIII., 12 = Paris (Babelon, Rois dc Syrie, p. 28, No. 207).

8. Similar ; style again less refined ;
features larger and coarser.

PI. XIII., 13 = Berlin (Prokesch)
(Bunbury).

B A Z I A E Q Z Similar type ; to 1., be-
A N T I o X o Y yond inscr., one-handled
vase r. ; beneath, -4 1 ail( l $ '■> Z3
to 1. of lower end of club, A .
Berlin (Lobbecke) = Num. Chron. 1883, PL iv. 4

9. Similar; but features considerably
altered.

PI. XIII., 14 = Copenhagen.

10. Same die as No. 9.

Berlin (Prokesch).

BAZIAE2Z Similar type; style
A N T I o X o Y rather better ; to L, be-
yond inscr., one-handled vase 1. ;
beneath $ and $ ; to 1. of lower
end of club, IY1 .

BAZIAEQZ Similar type; to 1., be-
ANTIoXoY yond inscr., one-handled
vase r. and, below it, ITI (?) ; be-
neath, d).

This completes our list of the varieties included in Class A, and it may
be convenient to glance back for a moment and take stock of its more
prominent features. In view of the testimony adduced in dealing with No. 1,
the attribution of the whole of these tetradrachms to Cyme surely needs no
justification. The one-handled vase is clearly discernible on the reverses of
all save No. 5 and No. 7. And in each of those cases the character of the obverse
forbids us to separate the piece from that which immediately precedes it. 24
Besides, even as exceptions they can be readily accounted for : on No. 5 the
part of the field usually occupied by the mint-mark is off the Jian, and on
No. 7 it is double-striking that has rendered the symbol unrecognizable. Our
scrutiny of the dies has revealed eight different obverses and nine different
reverses. Taking the latter first, we may note that the average level of
execution is decidedly high. There is none that is not at least passably good.
Further, they fall into two groups corresponding to the subdivisions indicated
in the list. Nos. 2 and 3, which belong to Subdivision (a), are ultimately
connected by the identity of the monograms that they bear, while other details,

- 3 The precise details of this monogram are a

little doubtful. They are quite clear on No. 9.

w The obverses of Nos. 4 and 5 (.Plate XIII.,

9 ami 10) are almost certainly from the >ann
hand ; and so with Nos. o" and 7 (I'laik XIII.,
11 and 12).

152 GEORGE MACDONALD

such as the placing of the symbol in the exergue and the attempt to indicate
the ground, suggest that both are from the hand of the same skilled
engraver. The remainder of the reverses present analogous points of resem-
blance not merely to one another but also to the coin of Antiochus I. Heracles,
for instance, wears a taenia, and the mint-mark, instead of being placed in the
exergue, is put in the field 1. The style varies too much to admit of our
assigning the whole set to a single engraver, but it is safe to say that one of
them has been a model for the rest. It has to be added that, while the bulk
of the monograms undoubtedly denote magistrates' names, there are two of
tli em to which a special character seems to attach ; some form of <fe occurs
on every reverse die from No. 1 to No. 10, and A appears on Nos. 2, 3, 4. 6,
and 8. 25

If we turn now to the obverses, we are impressed with the large propor-
tion of work that is almost first-rate. Nos. 2 and 3, which belong to Sub-
division (a) (Plate XII., 7 and 8), and Nos. 4 and 5, which stand at the head
of Subdivision (/3) (Plate XIII., 9 and 10), might all have served as ' patterns.'
The restraint and refinement they display are very noticeable. A coarser
touch obtrudes itself in Nos. 6 and 7 (Plate XIII., 11 and 12), and this
becomes more pronounced in No. 8 (Plate XIII., 13). It extends even to
the dots that form the border. Finally, on the die that is used for the obverse
of No. 9 (Plate XIII., 14) and No. 10, the size of the neck and chin is
suddenly so much reduced that the whole cast of the young king's features
undergoes a change. One is almost tempted to think that the engraver of
this die must have had before him — in addition to No. 7 or No. 8 — one of
the prototypes or 'pattern' pieces belonging to Class C (Plate XIV., 10f.),
and must have endeavoured to reconcile what seemed to him to be conflicting
likenesses. At all events, a survey of Class A as arranged upon Plate
XIII. discloses a great contrast between its two extremes. Yet the declension
is not nearly so rapid or so striking as in either of the two classes that are
still to come.

Antiochus II.

Class B. — Struck at Myrina.

Subdivision (a).

11. Head of Antiochus II., r., dia-
demed ; border of dots.

PI. XIV., A, -B.M.C. p. 15 No. 9.

12. Similar type ; style much coarser;
dots in border larger.

BAZIAEQZ Heracles seated 1. on
ANTloXoy rock, as on No. 1; to

1., beyond inscr., amphora ; to r. of

lower part of rock, fa.

Similar, with ex. line; style much
coarser; no monogram to r. of rock
but in ex., ft .
PI. XIV., 5=B.M.C. p. 15, No. 10 (Whittall); St. Petersburg (Hermitage).

25 Its absence from Nos. 5 and 7 may be part of the field is off the flan.
only apparent. In both cases a considerable

EARLY SELEUCID PORTRAITS.

153

13. Head of Antiochus II
denied ; border of dots.

Petrowicz Coll.

Subdivision (ft).

dia- BAIIAEQZ Heracles seated 1. on
ANTloXoY rock, as on No. 1; to
i\, beyond inscr., amphora ; be-
neath, jji and head of spear (or
arrow) r.
Num. Chron., 1883, PI. iv. 5 (Bunbury) ; Vienna.

14. Same die as No. 13. | Similar.

PI. XIV., 6= Leake, Nwm. Hell. Suppl. p. 4.- 15

15. Similar ; style much inferior. | Similar.

P.M. (Whittall) ; Berlin (Lijbbecke).

16. Same die as No. 15.

PI. XIV., 7 = B.M.

Similar.

Same die as No. 16.

17. Similar; head larger; features
varied.
PI. XIV., 8 = Cambridge (McClean) ; Berlin (Prokesch).

IS. Same die as No. 17. | Similar.

PI. XIV., 9 = Berlin ; Paris (Babelon, Eois de Syrie, p. 28, No. 208).

It will be observed that the link between the various members of
Class B is the presence of an amphora in the field. The analogy with
Class A makes it natural to interpret this as a mint-mark, and the clue thus
given is easy to follow up. At Myrina a similar amphora was used as a symbol
on tetradrachms with the types of Alexander the Great.' 27 It occurs in a like
capacity on the large ' spread ' pieces issued during the second century B.C.,
as will be seen from the fine specimen figured on Plate XIV"., 2. It was,
in fact, the ' town-arms ' of Myrina, 2S and as such it supplies a reverse type
for bronze coins of the third and second centuries B.C. : witness the example
reproduced on Plate XIV., 1. It is true that an amphora is a more common
numismatic object than a one-handled vase. The attribution of Class B to
Myrina may, therefore, for the moment seem less certain than was that of
Class A to Cyme. But the fullest confirmation will be forthcoming presently.

The subdivision of Class B into two sections was suggested by an exami-
nation of the coins themselves ; it is evident that there were two ' pattern '
pieces of somewhat different styles, and that each of these was made the basis
of imitations. Subdivision (a) contains only two varieties — the prototype and
an inferior copy. That this was the relation between them will, I think, be

:ti When Leake secured his electrotype, this
coin was in the Borrell Collection. I have
been unable to ascertain its present whereabouts.

'-" Midler. No: 933, ff.

88 See Wroth, B.M.C. Troas, Aeolis, and
Lesbos, p. lvi. Cf. supra, p. 147.

154 GEORGE MACDONALD

conceded by any one who looks carefully at the obverses as they are shown on
Plate XIV., 4 and 5. The testimony of the reverses is less clear. The
presence of the exergual line and of the magistrate's monogram beneath it,
as well as the absence of $ji, proves that the reverse of No. 12 — the execution
of which, by the way, is particularly coarse — was not modelled upon No. 11
alone. The engraver had also before him one of the coins belonging to
Class C ; compare, for instance, Plate XIV., 11. At the same time he did
keep No. 11 in view, as is clear from the position of the amphora in the field
and from the fact that it is a rock on which Heracles sits. 29 It is worth
noting that No. 11 bears no local magistrate's signature at all, for (h cannot
be a local monogram, seeing that it is found on eveiyone of the corresponding
coins from Cyme.

The monogram just mentioned is prominent on all the coins belonging to
Subdivision (/3) at Myrina. The latter comprises six distinct varieties,
including five reverses. The intimate connection between these reverses
does not admit of doubt. Monogram and mint-mark always occupy the same
position, and beside the monogram there is always a spear-head pointing
towards the r. The last-mentioned feature is at first sight rather puzzling.
Six made it a reason for assigning the coins to Sardis. 30 All becomes plain,
however, if we realize that it is not a mint-mark but a magistrate's symbol.
In fact, if we so interpret it, we get the promised confirmation of our
attribution of Class B to Myrina. The oldest known coins of this city are
small silver pieces of a high degree of rarity. The following is a description
of one now in the Fitzwilliam Museum at Cambridge; it weighs 1*88
grammes.

Head of Athena r., wearing Corinthian M Y Head of Artemis, three-quarter
helmet. face towards 1., wearing earring

and necklace ; quiver at r. shoul-
der ; in field r., head of spear (or

arrow).
PI. XIV., 3 = Leake, Num. Hell. p. 85.

The British Museum possesses one, the Berlin Museum two — a precisely
similar piece (from the Imhoof Collection) weighing 150 grammes, and
another (from the Lobbecke Collection) weighing 1*75 grammes and
differing from the preceding only in the absence of the quiver. 31 These
coins have been assigned, on .grounds of style, to the early part of the
third century B.C. If my interpretation of the spear-head be correct,
it enables us to date them more definitely still {circa 261 B.C.), and

2a See infra, pp. 156 f. Imhoof-Blumer, Z. f. N. xx. p. 282, where the

30 Num. Chron. 1898, p. 233. His view was Lobbecke coin is figured (PL x. 21). There is

that the whole of the Heracles tetradrachms another — according to Imhoof, a slightly earlier

were struck in the Sardian mint, but that they — group with the same types, but without the

also liore the symbols of some of the cities in spear-head ; two specimens described in Z.f. N.

which they were intended to circulate, such as iii. pp. 321 f., weigh 1*85 and 1 "80 grammes

Cyme and Phocaea. respectively; cp. B.M.C. ibid. No. 1.
81 See B.M.C. Lycia, etc., p. 69, No. 2;

EARLY SELEUCID PORTRAITS. 155

also provides us with the parallel which was spoken of above in discussing
the curious but unmistakable link between autonomous and regal money at
Cyme. 82 If the Cymaean coins are Persic didrachms, these may be
Persic diobols.

There is little more to be said about the reverses belonging to Subdivision
(ft). They are all fairly well executed, perhaps the least satisfactory being
that which is associated in No. 14 with the 'pattern' obverse. But they
present one point of considerable technical interest. In Nos. 15 and 16
we have an example of the same reverse die combined with two different
obverses. This is an inversion of the rule that is general here and elsewhere
in the Seleucid series. 33 An explanation may possibly be found in the compara-
tively low relief of the obverses concerned : the dies would be more liable to
breakage. Be that as it may, the circumstance is important as establishing an
intimate connection between two portraits that we might otherwise have
suspected of representing different individuals (Plate XIV., 7 and 8). We
may now safely attribute the lack of resemblance between them to the fact
that both are somewhat clumsy copies of the ' pattern ' (Plate XIV., 6). As
for the ' pattern ' itself, it is — like the corresponding head in Sub-division (a)
— quite a creditable piece of work. The artists who cut those dies had a
conception of the features of Antiochus II. that does not differ markedly
from the ideal expressed on the best executed pieces of Class A, and that is
also easily reconcilable with the more realistic ' standard likeness ' discussed
in my former paper. 34 In this respect they form a contrast to the engraver
whose work has next to be described.

Antiochus II.
Glass C. — Struck at Phocaea.

19. Head of Antiochus II., r., dia
denied ; border of dots.

B A 1 1 A E Q I Heracles 1. as in No. 1 ;

ANTloXoy but his seat, instead of
being a rough square rock, is
shaped like a tub or cauldron ; in
ex., £ and H\ 35
PI. XIV., 10 = B.M.C. p. 8, No. 2 ('Antiochus I.').

20. Similar. | Similar.

PI. XIV., 11 = Paris (Babelon, Eois dc Syrie, p. 28, No. 209).

21. Same die as No. 20. | Similar.

Petrowicz Coll. = Bunbury Sale Catalogue, Second Portion, Lot 454 (PL iv .

22. Similar; of much inferior style; BAIIAEQZ Similar type ; of much

ANT I oXoY inferior style; no taenia
visible ; in ex., A and 5 •

dots in border larger.

PI. XIV., 12 = Vienna.

a - See supra, p. 148. a5 These monograms are only partially visible.

B3 See J. U.S. x'xiii. pp. 99 f. But they can easily be completed from the two

84 J.H.S. xxiii. Plate I., 3 and 5. pieces that follow.

156 GEORGE MACDONALD

23. Similar ; style still farther dete-
riorated.

PI. XIV., 13= Berlin (Fox).

BAZIAEQZ Similar type; taenia
A NT I ox oy doubtful; ex. blank;
to 1., beyond inscr., head and neck
of griffin 1. and, below it, A.

24. Same die as No. 23. I Similar.
Berlin (Imhoof, Monn. gfeeq., p. 426, No. 29). 36

25. Same die as Nos. 23/. 1 Similar.
Paris (Babelon, Rois dc Syrie, p. 29, No. 210).

It will be observed that there is no common mint-mark to bind together
the members of Class C. And yet its homogeneity is beyond all shadow of
question. The seven varieties it includes involve four different obverses. The
first two of these (Plate XIV., 10 and 11) appear to be from the same hand.
They are so nearly alike that it requires a close examination to distinguish'
them. They are remarkable for the excellent workmanship they display, and
also for the peculiar characteristics of the portrait they present. The deeply
sunk eye and the long straight nose give the face a look that we do not usually
associate with Antiochus II. So much so is this the case that Gardner in
his Seleucid Kings felt justified in assigning No. 19 to Antiochus Soter. 37 '
His proposal met with opposition from the outset, 38 and it has been finally
disposed of through the discovery of No. 1. Its rejection leaves us with the
alternative of seeking the necessary explanation in the idiosyncrasy of the
artist ; but if he has erred, he deserves to be forgiven, for he has produced a
striking head. No. 22, though of much inferior style, is of quite exceptional
interest. No obverse in the whole group under discussion bears the stamp of
imitation in such an unmistakable degree. Its position on Plate XIV.
(12) is well adapted to bring this out ; it is plainly a crude attempt to
reproduce the head immediately above it. The mechanical treatment of the
loose ends of the diadem is specially significant. Of the die that served for
the obverses of Nos. 22 ff. (Plate XIV, 13) there is little or nothing to be said - r
it illustrates a still lower stage of degradation.

It is, however, to the reverses that we must turn for the most convincing
evidence of the homogeneous character of Class C. They number seven in all,
three (Nos. 19-21) of good workmanship, and the rest (Nos. 22-25) very much
inferior. Here, just as in the case of the obverses, it is evident that the
engraver of the ' pattern ' piece or pieces had ideas of his own. Heracles is-
not seated upon a conventional rock, as in Classes A and B, but upon an
object that has some resemblance to a tub or cauldron ; and the lower part of
the field of the coin is cut off by an exergual line. These features are faith-

38 Imhoof described the symbol, which is 37 Op. cit. p. xv.

obscure, as 'un buste de ccrf.' 38 jy unu Chron. 1883, p. 22 and p. 78.

EARLY SELEUCID PORTRAITS. 157

fully reproduced on the imitations. I do not think any mysterious significance
attaches to the former. I doubt whether it is meant to be more than a con-
veniently rounded stone. Certainly, if it is a tub or a cauldron, it must be
supposed to have a very stout lid ; Heracles is sitting well back towards the
centre. Ottfried Miiller called it a ' Kcssel,' so and Babelon and Six have
spoken of it as a ' cuve.' i0 The two" first-named saw in it an allusion to
the cleansing of the Augean stables. Whatever its true character may be, its
<jhief value for us is the function it discharges in holding together the
varieties we have grouped under the heading of Class C. It is the strongest
of the several indications that point so conclusively to a common origin.

And we can even determine the mint. Although the engraver or

O ©

engravers of the 'pattern' pieces did not deem it necessary to denote the
issuing city by a symbol, the influence of fashion was apparently too much for
their successors. On the last three reverses (Nos. 23-25), in the field beyond
the inscription — the very position so often occupied by the one-handled vase
at Cyme and by the amphora at Myrina, — we find the head and neck of a
griffin looking to the 1. We need not hesitate to interpret this as the mint-
mark of Phocaea in Ionia, the first important town on the coast to the south
of Cyme. The griffin, either in whole or in part, was a popular coin-type there
from the sixth century B.C. onwards; the reverse type of the fourth century
bronze piece which is figured on Plate XIII., 3 is exactly the same as the
symbol on Mos. 23 ff. The analogy with Cyme and Myrina is so complete that
no further proof appears to be required. I am tempted, however, to put forward
what seems to me to be an additional confirmation; if my suggestion is accepted,
■one hitherto unintelligible feature of Classes A and B will be satisfactorily
explained. It will be recollected that some form of the monogram <fo was of
practically universal occurrence on the reverses of the Heracles tetradrachms
struck at Cyme and at Myrina. Is it not probable that it represents the first
two letters of QwKaecov, and that it was placed upon the coins of the other
towns in token of the alliance that found expression in the issue ? This con-
jecture receives considerable support from the fact that the monogram actually
was, though to a less extent than the griffin, an acknowledged badge of
Phocaea. It is even employed by itself as a reverse type, one of the surest
indications of a ivapaa^fxov ; 41 see, for instance, the third century bronze piece
illustrated on Plate XIII., 4. And it appears on Alexandrine tetradrachms
of the city as a mint-mark, both alone and in company with a seal. 4 ' 2 Whether
any of the remaining monograms used in our group have an analogous signi-
ficance, may be doubted. But we may note in passing that A, which is as
unknown to Class B as is ^j to Class C, occurs on four out of seven reverses
at Phocaea, and on at least five out of nine at Cyme.

Our survey of the separate classes being thus completed, it only remains

39 Denhmaeler dcr alten Kunst, i. No. 236. 4J Miiller, Nos. 983 and 989. Others, also of

40 Lcs rois dc Syrie; p. lxi. ; and Num. Phocaea, show the forepart or the head of a
Chron. 1898, p. 233. griffin.

41 Of. Coin Types, pp. 123 ff.

158 GEORGE MACDONALD

to draw attention to certain general conclusions. It seems clear that about
261 B.C. something of the nature of a federal union subsisted between Cyme,
Myrina, and Phocaea. Cyme had minted no money since about 500 B.C.,
Myrina had never had a mint at all, Phocaea had issued neither gold nor
silver nor electrum since the expedition of Alexander the Great. Now, pro-
bably simultaneously, 43 the three cities begin to strike Seleucid tetradrachms
with a reverse type of a new and very remarkable character. And just about
the same period autonomous silver makes its appearance for a brief space at
Cyme and at Myrina. The significance of these circumstances cannot be mis-
interpreted. They point to common action on the part of the three towns
under the aegis of the Seleucid monarchy, action too that must have had a
successful issue, for it was not when defeats had been experienced that Greek
states took to striking money. It is not possible to conjecture with any degree
of confidence against whom the efforts of the league may have been directe:!.
Perhaps it was Pergamene or Egyptian aggression that had to be repelled.
Or we may have here an echo of that struggle against the Galatai that won
for Antiochus I. his honourable title of ' Saviour.'

And this may help us to see a fresh significance in the figure of the
resting Heracles. The type was subsequently imitated by Euthydemus I. of
Bactria (222-187 B.C.). It is found also on Spartan tetradrachms, some of them
autonomous, others bearing the name of the tyrant Nabis (207-192 B.C.).
Whether it has any special meaning on these, we cannot tell; it may be
merely an imitation. But, so far as regards the Seleucid tetradrachms,
some other explanation is required, and the most plausible hitherto available
has been the ingenious suggestion of Babelon. 44 He connected the sudden
appearance of Heracles with the dominance of Aristos and Themison,
the Cyprian brothers who were boon companions of Antiochus II. and to
whom that king surrendered much of his own power. 45 Pythermos describes
Themison as masquerading in a lion's skin with a club and bow, and allowing
himself to be hailed at festivals as Se/LLtacov Ma/ceScov. 'AvnoXov (3aai\e(i)s
'WpaKkr)<i. iQ With the discovery of No. 1 this explanation falls to the ground ;
the type is older than the reign of Antiochus II. Is it going too far afield
to recall the analogy of the anti-Spartan league of circa 394 B.C., when
Byzantium, Ephesus, Samos, Iasos, Rhodes, and Cnidus banded themselves
together in the cause of liberty ? 47 This latter confederation chose as its
characteristic coin-type a representation of the infant Heracles strangling the
serpents that threatened his destruction — a fitting enough symbol of a resolu-
tion to throw off a yoke that had grown intolerable. Is it not conceivable
that the Heracles at rest may have been meant to commemorate some great
struggle that had been brought to a successful conclusion ?

43 Although only Cyme is definitely known i. p. 336). Cf. Aelian. Var. Hist. ii. I.

to have minted with the head of Antiochus I., 4(i Apud Athen. vii. 289 f. (F.H.G. iv.

it is quite likely that the other two cities did 488).

tin- same. 47 See Waddington, Rev. Num. 1863, 223 ff.,

u Jcois rtc Hyric, p. lxi. and (for the most recent discussion of the

45 Phylarchos, apud Athen. x. 438 d. {F.H.G. subject) Eegling in Z.f.N. xxv. pp. 207 ff.

EARLY SELEUCID PORTRAITS. 159

The pages of written history have been scanned in vain for any
allusion to the anti-Spartan league of 394 B.C. ; the evidence for its existence
remains solely numismatic. In the case of the alliance which our study of the
Heracles tetradrachms has brought to light, a similar search has not proved
so absolutely barren. In his description of the successful campaign waged
by Attalus against Achaeus in 218 B.C., Polybius (v. 77) 48 thus opens his
account of the manner in which the tide of public opinion in the Hellenic
cities turned in favour of Pergamum : v Haav B)j Tore ixeTadep-evcu 7rpo?
avTov, irpMTov fiev Kv/xtj teal Ifivpva koX Qtotccua ' p.era 8e raura<i k.t.X.
That is the traditional text. Wilcken, however, has already pointed out that
S/J,vpva is an obvious corruption for Mvpiva. i0 His grounds for proposing
the change are twofold. In the first place, it would have been geographically
absurd to have ' sandwiched ' Smyrna between Cyme and Fhocaea. In the
second place, the participle /neraOepevai would have been quite inapplicable
to the conduct of the Smyrnaeans, who had successfully resisted both the
threats and the blandishments of Achaeus; a few lines further down
Polybius goes on to tell how Attalus i^ptj^dricre rots irapa tcov ^Lp,vpvalu>v
7rpecr/3evTai<; (pi\avdpoo7rco$, Sta to fxaXcara tovtovs reryprjKevaL tiju rrrpo?
civtov TTio-Ttv. Wilcken's emendation has been generally accepted and
hardly requires the additional support it now receives from the knowledge
that, forty or fifty years before the events narrated by Polybius, the three
towns had concluded a formal alliance. How long that alliance may have
endured we cannot say. But it would at least appear that in the crisis of
218 B.C. the memory of 261 was still sufficiently strong to ensure joint
political action.

I shall conclude by noting one other point where the numismatic

evidence we have marshalled can be brought to bear upon history. This

time its value is negative. Like so much else that happened in the third

century B.C., the earlier stages in the growth of the Pergamene kingdom are

wrapped in considerable obscurity. Strabo (XIII. p. 624) mentions that

Eumenes I. inflicted a heavy defeat on Antiochus I. at Sardis. Modern historians

have assumed that his victory was the signal for a wide extension of the

boundaries of Pergamum. It may have been so. But not all the inferences

based upon the passage can be justified. Niese 50 tentatively and Beloch ol with

much more confidence have assigned to that, date the boundary stone inscribed

opoi Uepya/jLi]i'coi>, which was found between Cyme and Myrina, 62 and which

must therefore have been erected after the territory of the latter city had

been incorporated in the dominions of Pergamum. Our coins forbid the

entertaining of any such suggestion. They show that at all events during

the earlier years of the reign of Antiochus II. Myrina still continued to

acknowledge the suzerainty of the Seleucidae.

George Macdonald.

43 I am indebted to Mr. EL L. Bevau for !0 Geseh. der Grieck. undMaked. -

directing my attention to this passage, p. 85, footnote '<.

« S.r. -Attalos' in Paulv-Wissowa, /,' M Gricch. Gcsch. iii. (i) p. 614, footnote 2.

JSneycl iii. 2162. s2 B. C. E. v. p.288.

INSTRUMENTAL MUSIC IN THE ROMAN AGE.

The cultivation of instrumental music remained in a backward state
among the Greeks in the fifth and fourth centuries, B.C. This was certainly
not due to any want of taste for music as a whole, for no race ever valued it
higher than did the Greeks. The reasons seem to have been, first, the
bondage of the instrument to the voice, second, the unsettled state of the
musical scale, and thirdly, the dislike of the Greeks for over-elaboration in
music.

These three points are well illustrated in the Republic of Plato. In
opening the discussion on the admissibility of certain modes, Socrates is
made to say that a musical composition is made up of three things, the
'words,' the 'harmony,' and the 'rhythm,' and that the musical ; words' are in
themselves in no way different from the words of common speech. 2 In
another place much scorn is cast by Glaucon on the musicians that sought for
the least perceptible interval to make that the unit of sound-measurement,
some of the experimenters declaring that they could distinguish an intermed-
iate note, where others insisted that the two sounds had passed into unison. 3
Socrates answers ' You mean those gentlemen who tease and torture the
strings, screwing them up on the pegs.'

Socrates and Glaucon both speak as amateurs in music, and their
feelings must have been shared by many Athenians at the time. Just as
with us there are some who long for the return of Handel's ' noble harmonies,'
as a relief from the chromatic aberrations of the Wagnerian school, so these
two worthy Greeks looked back to the sturdy Dorian airs of Terpander as
the true strains of the Hellenic muse, before she had learnt to voice the
subtler moods of the heart of man. To such amateurs the refinements
of the musical scale must have seemed base loans from the decadent

1 I wish to thank the following gentlemen 4 ; and Dr. W. H. D. Rouse for supplying me

for their land and valued help to me in with a modern example of a pan-pipe from

Collecting the materials for this article : M. Smyrna.

A. Merlin, Head of the Department of With very few exceptions, the monuments

Antiquities lor Tunisia, for having specially referred to are known to me in the originals,

had photographed for me the unpublished from my visits to the museums where they are

statuette from the Musee Alaoui, Tunis (Fig. to bo found.

5) as well as for much other assistance; - iii. 398 D. Cf. Nettleship, Lectures on

M. Gouvet, Director of the Museum at Susa Plato s Rep. 108.

(Sousse) for the prints used in Figures 2 and :; vii. 531 a.

INSTRUMENTAL MUSIC IN THE ROMAN AGE. 161

art of the East ; and if, as is now thought, the music of the fourth century
used intervals of a quarter, three-eighths, a third, two-thirds, and three-
quarters of a tone, this was no wonder. 4

The contrast is drawn by Socrates between the school to which Aristo-
xenus belonged, the cultivators of the enharmonic style, and the Pythagoreans,
who based the scale on the harmonic relations of the octave, fourth, and fifth.
Into the details of this controversy there is no need to enter, but it is clear
that such a simple and easily-tuned instrument as the lyre was best fitted to
this screwing-up process by which the minimum intervals were reached. A
many-stringed instrument would have taken too long to adjust to any
highly complicated system. By the time of Aristoxenus himself the
enharmonic scale was nearly dead, as Aristoxenus himself regrets;"' and
although the later musical writers repeat mechanically their account of it,
there is not much reason for thinking that it was ever revived in practice.
Aristoxenus complains that, if the enharmonic system was dropped, there would
soon be nothing left but the diatonic and the ' highstrung ' chromatic {^pwyni
(tvvtovov or rovialov 6 ), and these actually survived.

The objection felt by many Greeks to variety of musical effect is voiced
by Socrates soon after his remark first quoted. He banishes such many-
stringed and various instruments as the 'Triangle,' the ' Pectis,' and all kinds
of flutes," leaving only the lyre, cithara, and, for shepherds, the pan-pipe
{syrinx). It is possible that in retaining the cithara, Socrates may have
meant only the kind with few strings, for it would have been strange to
admit a fourteen-stringed cithara, while condemning the flute for its too
great variety of sound. Here Plato's views must have seemed very narrow
even to his own age. In making music a means of moral upbuilding he not
only struck at virtuosity and over-refinement, but would have checked the
progress of the art along its most promising lines. His beliefs do not seem to
have had much effect, for the very instruments that he excluded were culti-
vated with growing zeal. In Greece itself however the double-flute, lyre, and
cithara remained the favourites. At Athens every boy was taught the lyre, ami

4 Cf. WestphaL Harm. u. .1/./. -/. Gr. 4'. 17. chromatic could b>- played on a piano; and

Westphal's view holds the field, but it ne.-ds although Ptolemy has three kinds of diatonic

some t'aitli to believe that these ear-splitting scale, it would -till seem that the music of his

dissonances were commonly played and sung. day would not have sounded utterly bai

The so-called enharmonic mode of the modern to our ears. A form of chromatic mode is in

Eastern Church CHxos rp'nos ; cf. I. Th. use in the Eastern Church, and is often heard

Sakellarides 'Upk 'r/xv^Sla 95) is sung like in Romaic folk-songs: it has an austere and

thr major scale of F; and it lias been supposed striking effect. A -soli ' diatonic is .sung in

by Doin (iaisser La Musique Eccl. Or. 0' apres -ome Greek churches as the second Byzantine

la Tradition) thai the ancient enharmonic was mode; but few western listeners find much

the same as this, the two quarter- ss in it. We ourselves allow both the

tones being always sung together, and the '.just' and the 'tempered' intonation; the

double tone being divided. It is impossible bagpipes, I believe, are tuned to neither of

now to go into this interesting theory. these, and their effect is not always dislik

Quoted in Pint, de Mus. 38. Cf. Aristox. 7 iii. 399 C and i>. Aim. Pol. viii. will not

Harm. i. 23. allow these instruments in the training of th.'

8 This is stated by Ptolemy. Cf. Monro, young.
Modes of Ancient Gfr. Mus. 111. The highstrung

U.S. — VOL. XXVII. M

162

H. J. W. TILLYARD

the use of the flute was by no means confined to professionals : Alcibiades, for
example, is said to have studied on it. s Again Epaminondas not only played
the flute like other Thebans, but learnt the lyre also. The cithara, which
had been perfected by Timotheus, 9 was chiefly played by professionals. These
three instruments are common subjects on Attic vases, and late monuments
and authorities show that their use went on through the Roman age : their
nature however is so well known that there is no need to say more about
them.

The kinds of harp called ' Triangle ' and ' Pectis ' by Plato seem to have
been of Lydian origin. Athenaeus, who has a long discussion on the subject,
says that the Magadis was a stringed instrument, later called Sambuca, while
the Pectis was the same. 10 It is possible that the ' Triangle ' was also similar,
and that the names of Trigonon and Pectis were meant as Greek renderings
for the foreign words Magadis and Sambuca. This harp, as it may safely be
called, is often seen in Egyptian art, and must have been widespread over the
East. It appears on a fine red-figured vase in conjunction with the lyre and

cithara. 11 Athenaeus says that Sappho
brought in its use from Lydia, and Anacreon,
as his own words record, played a harp with
twenty strings. 12 The instrument embraced
the whole compass of the singing voice, and
had high notes beyond the range of the
cithara. 13 It could be used without a
striker. One of its peculiarities was that
the sound-box was on the upper side.
Smaller sizes with nine or even five strings
were sometimes made. Examples of such miniature harps are seen in some
of the wall-paintings now in the Naples Museum : one of these, played by a
Cupid, is here illustrated. (Fig. l) u

Another stringed instrument of the same class, more like a zither, became
popular in the Roman Age. But, while the use of the harp called for great
skill, and gave full scope for rich and splendid effects, the zither can only
have yielded a thin and twanging tone, especially as the ancient instrument
often had no sound-box. This instrument is nearly always played by women,
who often wear a carelessly sumptuous dress, suited rather to paid performers
than to freewomen. Examples are again seen in Roman wall-paintings. In

Fig. 1. — Tp.igonum on Sambuca.

h Duris ap. Athen. iv. 84, 184 d, whore it is
said that the famous flute-player Pronomus was
Alcibiades' master.

! » Cf. Pans. iii. 12. 8.

'' Athenaeus xiv. 34-38. Aristotle I.e.
classes together the Pectis, Barbiton, Heptagon
(otherwise unknown), Triangle, and Sambuca:
he rails them apxc'ia vpyava.

11 Pauini'ister, Dculemaler, 1544.

12 Anacr. 14(5). \pd\\w 5' tlico<T(ixopoov)
<lv x*P^v> fw.ya.lw tx<if. In fr.l3(I6) he

speaks of the Pectis.

13 Telestes 4 (5) (Bergk) rol 8' o^vcpwvois
TrriKTtSccv ^aA/AO? < s> KpeKov AvStov vjjlvov.

14 For other examples cf. Le Pitturc Ant. d'
Ercolcmo. v. 167. Inghirami, Pitt, cli Vasi
Fittili iv. cccxliii. , and von Jan Arch. Zeit.
xvi. 187 (PI. cxv. 14). For these instruments
in general cf. an article by von Jan in
Baumeister s.r. Sail en instrumental ; and a
dissertation of the same writer, Die gr.
Saiteninstrumenten.

INSTRUMENTAL MUSIC IX THE ROMAN AGE.

163

Roman Africa a complicated form of this instrument was common, and it is
often represented in statuettes. The example (Fig. 2) is from Susa (Had-
rumetum) and shows a lady performer, richly dressed, and possibly wearing
a wig : on either side of her is a small figure,
perhaps of a muse. The type is often seen in
Africa. 15 The name of this zither was perhaps
the Psaltery.

A more curious contrivance was the
so-called Tripod, invented by Pythagoras of
Zacynthus, and described by Athenaeus. 1 ' 1
This stood on a revolving base with a sound-
box called A,e/3?7? above ; the strings were
stretched between the three branches that
gave the instrument its name. On one side
the strings were tuned in the Dorian mode, on
another in the Phrygian, and on the third in
the Lydian. If the player wished to change
the mode, he had only to turn the instrument
with his foot, so as to bring another row of
strings within his reach. The left hand was
used to stop down the strings or to check
their vibration, 17 and the right hand held the
striker. This ingenious instrument did not
survive the death of its inventor.

The principle of shortening a string to
make higher notes was known to the Greeks
at an early age. Nicomachus says that the
Pythagoreans called a one-stringed lute a Canon, 18 which means that it
was used by them as a standard for generating their scale. It is quite
possible that the range of the cithara was sometimes extended by stopping
down the strings, but this was no part of the regular technique of that
instrument. In the Roman age instruments appear in which the strings were
systematically stopped down on a finger-board as in a modern mandoline.
Although it is possible to embrace a large compass of notes in this way, the
tone produced must always have been feeble and lacking in resonance ; and
as now the guitar, mandoline, ami banjo are hardly reckoned as instruments
of music, so the use of such instruments in old time was a sign of declining
taste. The ancient name seems to have been the Pandura. Pollux remark-
that the Pandura had three strings, ami was invented by the Assyrians. 19
Nicomachus, in the place already quoted, classes the Pandura with the one-

Fig. 2.— Terracotta Figure at

Susa. Woman with Zither
(Psaltkkium).

13 Cf. two similar figures in tin- Musde
Alaoui, Tunis {Miuie "'" Bardo 71, 7"> ; in the
series SfusA i tt Collections arch. d'Algir
hi Tunisie, ed. La Bl inch) re.)

'" A then. xiv. 41.

17 Aio rr)V ewifioXijr.

18 Harm. p. 8.

'' Pollux I-'. 60. Tf>i\op5oi- 5i $xcp 'Affa6pioi
irarbovpar atvofiafav, iKttwv 5' ij»' rb evp-q/ua.

Where however Dindorf notes that no form ol
the nann> Pandura is found either in Assyrian
or Chaldaean.

M 2

164

H. J. W. TILLY ARD

stringed lute. But this can hardly mean more than that the notes were
produced on the same principle, namely by stopping down the strings. Other
writers confuse the Pandura with the pan-pipe ; 20 but as the name is still
applied to a stringed instrument in Italy, it can hardly have been otherwise
in antiquity. Martianus Capella 21 calls it an Egyptian invention, and it is
probable at any rate that it came from the East. The Emperor Helioga-
balus, who was brought up in Syria, used, among his other undignified
pursuits, to play on the Pandura ; 22 and one is represented on a silver cup
of Graeco-Persian workmanship found in South Russia. 23

About a dozen examples of this type of instrument are known, and none
of them is earlier than the Roman age. 24 Two principal shapes may
be distinguished. One is shaped very much like a mandoline, with an
oval shell and a short neck. Of this the Graeco-Persian Pandura is a

specimen; and there is another played by a
siren on a sarcophagus in the Lateran
Museum, 25 and another, almost guitar-shaped,
in the museum at Turin (Fig. 3).* 25a The
other form resembles a banjo: it has a very
long neck, but instead of a drum head
stretched over a hoop, a round shell is used
to re-inforce the sound. The back and front of such instruments, with
the manner of playing, are shown in the illustrations (Figs. 4 and 5). These
are taken from African statuettes ; but the type is not at all common. The
other extant examples are chiefly on sarcophagi. A fine specimen is seen
in the representation of the wedding of Cupid and Psyche on a late
sarcophagus in the British Museum. 20 At Naples 27 there is a sarcophagus
which is remarkable because not only one of the figures in the scene repre-
sented, but also the lady who appears on the medallion, and was therefore
buried in the coffin, are playing the Pandura. The instruments here are shaped
like the African specimens, having a crescent-shaped top, and four strings
instead of the three mentioned by Pollux. It cannot be seen whether
the finger-board was divided by ridges, as in modern instruments of that class.
On the African statuette it would almost seem that the strings are stretched
over a bridge, but this also is uncertain.

Among the instruments condemned by Plato are all kinds of flutes.' 28

Fig. 3. — Pandura.

E.g. Isid. Orig. 3. 20.

21 9. 924. Aihenaeus iv. 82 ascribes its dis-

, t>> tin- Ti'i;_'lo<lytes by the Red Sea.

22 ' Panduriza/vit,' Lamprid. IIc.l. 32.

-' Stcphalii. (Jon, j,/' J.'.n'/n, 1881, 5ft.

24 There is a list given byStephani ib. Some
<1 Li- examples are doubtful. The supposed
Pandura on the well-known Hippolytus relief
on a fine early sarcophagus in the cathedral
ii Girgenti 'Arch. Zeit. 1847, PI. VI.) seems to
ne alter close inspection) to be only an
elongated lyre. The instrument on a relief in

the Louvre (Clarac, Mus. Sculp. 119, No. 47 ;
cf. Robert, Ant. Sark. ii. 41, pi. 26 a) is also
hardly a Pandura.

25 Benndorf, 126.

23:1 On a late relief of Orpheus and the
Nymphs. Unpublished, but possibly forged.

28 Ancient Marbles in B.M. PI. IX. Fig. 3,
ami p. 35.

27 Naples Museum, No. 6598.

28 Arist. I.e. also rejects the flute in educa-
tion.

INSTi; I MENTAL MUSIC IN THE ROMAN AGE. 1G5

He was no doubt thinking chiefly of the double-flute, which had reached
a high pitch of complication in his own day, 29 and which is often represented
on Attic vases. The instrument belonged rather to the flageolet class
and had a mouthpiece. The true flute-type (ir\ayiav\o<;) was also known
but little cultivated. It is sometimes seen as a short fife played by Fauns and
Satyrs; or by Cupids, as on the urn of L. Minucius Felix in the Capitol
Museum at Rome. 30 A more interesting type of wind-instrument, appearing
in the Roman age, had a wing-joint, and resembled a bassoon. This is

Fio. 4. — Terracotta Figure
at Svsa. Woman with
Pandura .and Wreath.

Fig. 5.

-Terkacotta Figure in the M usee Alaoui.

Max Playing ox the Panduiia.

seen on a sarcophagus in the Taormina Museum, hero illustrated (Figs. (>. 7).
It will be seen that the sculpture is late work. The heads of the figures are
too big, the iris of the eye is hollowed out, and the hair and drapery are
freely worked with the drill. The sarcophagus, which was meant fir a child,
may therefore date from the third contury A.D. Besides the bass wind-
instrument there are also a lyre, cymbals, a conch-shell, and a small pan-pipe
in use. The name of the instrument is uncertain : it may have been
the Bombalium. 31 There is, I believe, only one other example, which is seen
on a small sarcophagus in the Vatican. 32

29 It had been perfected l>\ Pronomus tin'
master of Alcibiades. Pans. ix. 12. 5.

30 H.ll.i^, Fuhrer, 440.

:I This word occurs Epith. Laurent. 61.

■'■- Mus. Pio-Clem. v. 1:3. It i> doubtful

whether the Pan in the British Museun
'•; B.Af. iii. 135 is playing such an

instrument. (Cf. the article in Baumeis
FlStcn.)

166

H. J. W. TILL YARD

The pan-pipe (Syrinx, Fig. 8) was one of the oldest Greek instruments, and
was always put in the hands of shepherds and country deities. In pastoral

4 W'$*** '

4 vt ?:

Fig. 6.— Sarcophagus in the Taobmina Museum.

Fig. 7.— Part of the Sarcophagus Shown in Fig 6.

poetry it is often mentioned, and it appears in art as an attribute of Pan and
of Fauns and Satyrs. It consisted of a row of pipes made of cane-stalk,
each pipe being stopped by the natural joint of the cane, below which the

INSTRUMENTAL MUSIC IN THE ROMAN AGE.

167

pipe was cut off. The pipes were put in a row, and, as Pollux says, 33
fastened together with thread and wax. Below and above the row of pipes
two flat strips of cane were laid, and to these the thread was tied, going
round the reeds and holding them firmly together. It is easy to see from
ancient sculpture that this was the plan then followed, and a modern
pan-pipe from Smyrna, now in my hands, has been put together in the same
way.

In the Greek pan-pipe, the reeds all appear of the same length : there
were as a rule about eight of these. 34 Such an instrument is played by

i 11 n II ii ii ii ii

nnnnnnnnm

Fig. 8.— Pan-pipes : {a) Greek, (6) Graeco-Roman.
(The natural joints of the cane are shown in black.)

Calliope on the Francois Vase, and it appears as an attribute of Pan on
Arcadian, Messenian, and Sicilian coins. 30 In later art it is rare, though
there is a good example on a relief of the Hellenistic age in the Barracco
Museum at Rome, representing Pan and the Nymphs.

As the reeds were all of the same length, how were the different notes
made ? Some have thought that there was a row of holes in the pipes at
certain heights above the joints ; but I have found that a hole in the side of
the pipe takes away the musical tone altogether. It is not likely in itself,
nor does it appear from the monuments that difference of thickness was the
sole basis of the scale. Probably therefore the reeds instead of being cut
off just below the joint of the cane, as in the modern pipe, were cut some
inches longer than they were meant to be, and when the upper part of each
had been trimmed to the length required, the lower ends were simply cut off
so as to leave the pipes even, although the part of each pipe above the joint
(which alone made the note) would be different in every case. The advantage
of this plan would be that reeds of the same length would be easier to
fasten together, and the pan-pipe thus made would be more handy to hold.

13 Pollux iv. 69 (ffvpty^) . . . i] fi-if ovv Ka\d-
uuic iffTt <xvv8T}Kri AtVy k<x\ Kriptji crvvSeduaa. The
pan-pipe is also an attribute of Attis. Cf. the
terracottas in B.C.K xxi. 518 5 - JU. The reeda
are all of the Bame length.

■ Cf. Ann. d. Inst. 1877, 211.
< p. Gard ns, PI. II. 12.

W. M Leake, K . Hell. 17. Head,Hist.

373.

168 H. J. W. TILL YARD

The Roman form of the pan-pipe was also the Etruscan 36 and the modern
shape : in this the reeds were cut off just below the joints of the cane, and
so bound together as to leave the ends of the instrument sloping. 37 This
became the recognised form in Graeco-Roman art, and it is very widespread
on the monuments. It seems to have taken the fancy of the mediaeval
restorers, so that countless statues have been embellished with pan-pipes
in plaster.

Pan-pipes were made in all sizes ; some had only four or five small
reeds ; one of these is seen on the sarcophagus from Taormina, already
illustrated ; some had as many as twelve reeds bound with three bands. 3S
A piece of ribbon was sometimes fastened to the instrument by which it
could be carried when not in use. The scale of the pan-pipe no doubt
varied with the maker's taste, but it was probably diatonic as a rule :
firstly because it would be the easiest to make in tune, and secondly
because the murmuring or buzzing effect produced by gliding from note
to note would have been harsh and dissonant on any other system. By
strengthening the blowing each reed could be made to yield a note
an octave higher than its normal pitch ; so that perhaps the instrument
with seven or eight reeds was strictly diatonic, while the more complex
and rarer kinds had chromatic notes in between. The shape with a double
row of pipes seems to have been invented by the restorers of statues in
the middle ages. 39

Literary references to the pan-pipe are very common : it will be enough
to quote a pretty description from Claudian, which shows how the instrument
was played. (Epith. Pall, et Cel. 34)

(Hymenaeus) . . . platano namque ille sub alta
fusus inaequales cera texebat avenas,
Maenaliosque modos et pastoralia labris
murmura tentabat relegens, orisque recursu
dissimuli tenuem variabat arundine ventum.

Besides the common pan-pipe which was blown from the top of the
reeds, the Romans invented a more complicated kind known as the Etrus-
can Pipe. Pollux explains that this was made of bronze reeds and was
played upside down, the smaller sort being blown by the breath. 40 Of
this instrument the pan-pipes now extant in the Naples Museum are ex-
amples. The larger sort, says Pollux, was blown by water ; so that it is
clear that he is referring in both cases to primitive kinds of organ with

38 It appears on an Etruscan urn, Brunn, is nearly all plaster.

Bilievi </. Urnc Etrus. 1, PI. 92, 3 ; cf. Jlu//. 40 Pollux iv. 70 toDto Se kclto. epimKiv ex uv *>

Inst. 1886, 1, 101. Tupprjvbs ai)\hs a.VTeaTpa/J.fj.€Vi) vvpiyynrapeoiKws,

'■'" Cf. Oviil. Met. i. 710, disparibus calamis. x a ^ K ^ s M e '" fffTtv 6 KaXa/xos, KaToodtv Se xnroirve6-

18 The pan pipe on the glass krater in the [xtvos. fixrats fx\v b £\d.TTwv, liScm 5e 6 ixsifav

Naples Museum {Cat. p. 91) lias twelve reeds. ava.B\ifiofx{v<p ica\ aftpav -nvivfxaios acpievri. For

'* Thus the example in the Vatican (Ann-- the instruments at Naples cf. C. Abdy Williams,

lxmg,[Sculp. d. Vat. Mus., M. Chiaramonti 588] Class. Rev. 1902, 409.

INSTRUMENTAL MUSIC IN THE ROMAN AGE. 169

some sort of mechanical fingering'; and it is curious to note that from the
pan-pipe which was deemed only good enough for shepherds should have
grown the most majestic of all instruments. It is well known that the
organ was highly developed in the later Roman age, 41 as may be inferred
from a fine passage of Claudian {Be Mall. Theod. Cons. 316).

Et qui magna levi detrudens murmura tactu
innumeras voces segetis moderatus aenae,
intonat erranti digito, penitusque trabali
vecte laborantes in carmina concitet undas.

It is remarkable that this noble instrument described by Claudian
served no better end than to amuse the crowd gathered in the amphitheatre,
where its strains alternated with the feats of tumblers and the sallies of
buffoons. But this after all may be characteristic of the music of the Roman
age : it was no longer dedicated chiefly to the worship of the gods, or to
the serious education of youth ; the advance in skill and in the variety
of instruments did not imply a real progress in the art, but rather led to
virtuosity and false effect.

H. J. W. TlLLYARD.

41 On the Roman organ v. Baumeister s.v. upon here.
Floten. The subject is too wide to be entered

A HISTORY OF THE PELASGIAN THEORY.

Few peoples of the ancient world have given rise to so much controversy
as the Pelasgians ; and of few, after some centuries of discussion, is so little
clearly established. Like the Phoenicians, the Celts, and of recent years the
Teutons, they have been a peg upon which to hang all sorts of speculation ;
and whenever an inconvenient circumstance has deranged the symmetry of a
theory, it has been safe to ' call it Pelasgian and pass on.'

One main reason for this ill-repute, into which the Pelasgian name has
fallen, has been the very uncritical fashion in which the ancient statements
about the Pelasgians have commonly been mishandled. It has been the
custom to treat passages from Homer, from Herodotus, from Ephorus, and
from Pausanias, as if they were so many interchangeable bricks to build up
the speculative edifice ; as if it needed no proof that genealogies found sum-
marized in Pausanias or Apollodorus ' were taken by them from poems of the
same class with the Thcogony, or from ancient treatises, or from prevalent
opinions ; ' as if, further, ' if we find them mentioning the Pelasgian nation,
they do at all events belong to an age when that name and people had
nothing of the mystery which they bore to the eyes of the later Greeks, for
instance of Strabo ; ' and as though (in the same passage) a statement of
.Stephanus of Byzantium about Pelasgians in Italy ' were evidence to the same
effect, perfectly unexceptionable and as strictly historical as the case will
admit of.' l

No one doubts, of course, either that popular tradition may transmit, or
that late writers may transcribe, statements which come from very early, and
even from contemporary sources. But this is quite a different matter from
assuming, as a working hypothesis, that the unauthenticated statements of
late writers do come from early sources. Even where such a statement tallies
with a statement of Homer, or with the results of excavation, we are not
justified in inferring, on that account only, that the late writer had Homer
before him, any more than that he had himself conducted such an excavation.
In the absence of evidence to the contrary, he may equally well be assumed
to have got his information from a quite late handbook, or from an imaginative
author who for once by chance was right.

Most recent writers meanwhile admit, tacitly, that authorities do vary

1 Niebuhr, History of Home (tr. Hare and Thirlwall 1837; i. p. 26.

A HISTORY OF THE PELASGIAN THEORY. 171

in value, and that ceteris paribus the earlier sources are more trustworthy
than the later. But the reservation 'ceteris paribus' covers a great deal; for
it is argued, not uncommonly, that Hecataeus, for example, stands much
nearer in the scale to Pausanias than he does to Hesiod, and Hesiod nearer
to Hecataeus than to Homer : in the sense, of course, that between Homel-
and Hesiod lies a great political convulsion, involving a fatal breach with the
past ; and that between Hesiod and Hecataeus lies at least a century of
strenuous endeavour to bridge that gap, and ' restore ' the missing data by
strenuous use of the imagination.

At the time when the chronological lacuna between Mycenaean and
Hellenic Greece was still unsurveyed, a considerable service was rendered by
Mr. Cecil Torr, in an experimental reconstruction, 111 in which every interval of
time which he was able to demonstrate was ' written down' (so to speak) to
the ' least possible ' dimensions; somewhat as if a prudent capitalist to-day
were to ' write down ' to SO the value of his consols. The result was a
chronological scheme which, although it has not been widely adopted, had at
least the merit of being ' within the mark.' It called attention, besides, to
certain other matters of historical method, which I need not specify here.

Now what I have attempted to do, in this essay, is to make a similar
experiment with the ancient statements about the Pelasgians: to arrange
them, in fact, strictly in accordance with the relative antiquity of the sources
from which they severally become first known to us ; and to use, at each
stage, as commentary upon any passage, only such other statements as we
know from extant authors to have been current at the date when that passage
was penned. To interpret Homeric passages, that is, I shall use only Homeric
evidence and the physique of the Aegean, accessible to ' Homer ' as to us ;
to interpret Hesiod and the later Epic, only Epic sources ; to interpret
Thucydides, only sources of at least fifth century date. Not until I reach
the authorities of the age of Alexander, shall I make use of any statement
which rests merely on the authority of Ephorus or his kind. In this way
alone, I think, can we be certain to avoid anachronism. Much else about the
Pelasgians may very likely be ancient tradition, but it cannot be proved from
extant sources to be so; and it may, on the other hand, find a more probable
context — if not an assured origin — lower down, when once we have constructed,
on the hypothesis of 'lowest possible' dates for each phase, the outlines of
the growth of the Pelasgian Theory.

It is difficult to be certain, in an enquiry of this kiud, that one has
really left preconceptions behind; but I may at all events confess this, that
I had not the faintest idea, when I began to apply this method to my
materials, what the results of the experiment were going to be. Least of all
was I prepared for the form which the Homeric evidence assumed, when once
it was released from its Hellenic commentary; or for the part which I have
found myself compelled to assign to Ephorus in the concoction of the Great
Pelasgian Myth.

1:1 C. Torr Memphis and Mycenae, Cambridge, 1896.

172 J. L. MYRES

§ 1. — Homeric Evidence: its Two-fold Character.

To take, first and separate! y^the Homeric passages. 11 ' They divide at
once into two classes : those which contain the substantival forms UeXaayos,
UeXacryoi and those which contain merely the adjective Ue\aayi/c6<?. In the
substantival passages it is a fair preliminary hypothesis that the poet had in
his mind some more or less definite conception of an actual people, either
still existent in his own time and that of his original audiences ; or, if
extinct, familiar both to him and to his audiences, through a lively and
accepted tradition, as recent occupants of the areas in which he places them.
In the adjectival passages, on the other hand, such a hypothesis is not
legitimate. These do not indicate more than that the place or personality to
which the poet applies the adjective 'Pelasgian' seemed to him, and presumably
to his audience, to partake, in some way, of the Pelasgian character as he or
they understand it. These passages therefore cannot be used by themselves as
evidence that either the audience or the poet had any experience or
immediate reminiscence of actual Pelasgian inhabitants in the area or about
the personage to which the adjective is applied. And when we come to
consider this class of passages in detail (B. below) we shall see, I think, that
this consideration is valid, and of some importance.

| 2. — Substantival UeXaayol in Homer.

It will simplify discussion to take the substantival passages first. They
are as follows : —

(1)
II. 2. 840-3 : '\tttt66oos 8' dye (pu\a TleXaaywv ey^ecrtpcopcov,
roiv o'l Adptaav epi/3oo\aKa vaierdaaKov

TWV ?]PX T7T7TO^OO? T€ UvXaiOS t' 6'£o9 "A/3^09,

vie &va) Aijdoio YleXacryov TevTap,i8ao.

The passage stands at a critical point in the structure of the 'Trojan
Catalogue.' Starting from Troy-Town, in 1. 816, the poet has reviewed (1)
the Trojans themselves (11. 816 ff.) ; (2) their Dardanian neighbours to the
N.E. (819 ff.) ; (3) other Trojans from Zeleia (823 ff.) on the lower Aisepos,
where the lowest spurs of Ida sink into the Propontic seaboard; (4) Adrasteia

Ul It might fairly be argued that account on the view that relatively — though of course

should 1»- taken hen of tin: possibility that not absolutely — these minor distinctions are

the Odyssey for example may represent a later unimportant ; and that even if some parts of

phase of Homeric belief or of Aegean history 'Homer' may possibly be approximately as

than the Iliad ; or that a distinction should be late as some parts of ' Hesiod,' clearness will be

observed between data supplied by the 'earlier' gained, without sacrifice of truth, by treating

or the 'later' parts of the Iliad. But, quite the Homeric Epic as a single group of data,

apart from the uncertainty which surrounds and Hesiod and the other fragments of Epic as

the whole question of such dissection of the a distinct, and on the whole well contrasted

Homeric corpus, I have thought it better to act group.

A HISTORY OF THE PELASGIAN THEORY. 173

(simply ' Adrastos' town,' like Midaeion, Kotyaion, and the like) with Paisos
(Apaisos), and Mt. Tereia, between Parion and Lauipsakos (835 If.) : i.e. the
poet has reached the E. margin of the Troad, and is returning by the sea-
coast to (5) Perkote, Praktios (river), Arisbe (on the Selleis river), Abydos, and
Sestos. With the mention of Sestos we have passed from Asia into Europe.
Then come the Pelasgians (1. 840): then (6) the Thracians, 'all those whose
frontier is the Hellespont ' 845 : then (7) the Kikones (11. 84-6 ff.), who are
fixed by Od. 9. 39-40 in their historic habitat ' under Ismaros,' west of the
lower Hebros : then (8) the Paeonians (11. 848 ff.), who come from as far off as
the Axios river. Here the confederacy of Priam has its limit westward ;
and the poet starts again from the Troad, and strikes out, first north-eastward
through Paphlagonia and beyond; and then finally southward, through
Mysia, Phrygia, Maeonia, and Caria, to Lycia, where the confederacy ends
south-eastward. Priam's confederacy, in fact, once plotted out upon the map,
reveals itself as a coalition of the whole northern and eastern shores of the
Aegean against a ' blow at the heart ' delivered by Agamemnon, as overlord
of the south and east from Kos and Rhodes to Olympus, Ithaca, and Dodona.
Now the whole of the rest of this tripartite list is in correct geographical
order so far as it goes ; and the single omission of importance (that of
Bithynia, between the Troad frontier at Zeleia on the Aisepos, and the
Paphlagonians) is sufficiently accounted for («) by the later consensus that
the historic Bithynians (like the Mygdones of the Odryses river, inland of
Daskyleion and Myrlea) were Thracians-in-Asia, whereas for the Catalogue-
poet the limit of Priam's Thracians is the Hellespont; 2 (b) by the indication
supplied by II. 3. 184 ff. that the Phrygians themselves were but recently
arrived in what later became Bithynia, and were still cutting their way up
the Sangarios valley in the early manhood of King Priam.

The Catalogue, then, sets a block of Pelasgians between the home-
country of the Troad and the Thracians ; and the mention of Sestos in the
previous section, along with Abydos and Arisbe, shows that the poet's survey
has already reached and crossed the Hellespont. The probability therefore is
that the Pelasgians of the Catalogue occupied an area between the Helles-
pont at Sestos, and the proper country of the Thracians.

At this point a geographical consideration comes to our aid. Between
the Isthmus of the Chersonese, and the headquarters of the Thracians in the
basin of the Hebrus, lies the rougher and more hilly tract from C. Sarpedon
to the Hieron Oros, which in historic times was occupied by the Caeni and
Apsinthians, but which, though overrun thus later by Thracian tribes, never
became wholly incorporated in the geographical area of ' Thrace.' It is
therefore not unreasonable to suppose that this same area corresponds with
the non-Thracian, and at the same time non-Hellespontine area, which the
poet of the Catalogue assigns to the ' Pelasgians.'

- In post-Homeric time we shall find copious Homeric evidence as commentary on II

evidence of this Thrako-Pbrygian thrusl south- but only because the event under discussion is

eastward across the Hellespontine area. I itself i post-Homeric,

have broken here my rule of not using post-

174 J. L. MYRES

It was inevitable that the occurrence of the place-name Larisa in this
passage should give rise to copious speculation : particularly as one of the
principal towns of Thessaly bore this name, and lay at no very great distance
either from ' Pelasgian ' Argos or from ' Pelasgian ' Zeus at Dodona ; and
another Larisa (L. Kremaste, not mentioned in Homer) lay later closely
adjacent to the former, in the territory assigned to Protesilaus. Prof.
Ridgeway, for example,' 2a pronounces without hesitation for the Thessalian
Larisa, and avoids the obvious difficulty, how people from the Thessalian
Larisa should be fighting on Priam's side, by laying stress on the form
vaLerdaaKov as meaning ' used to live there, but have ceased to live there
now.' But exactly the same grammatical form is used of the men of
Karystos and Styra (1. 539) ; and there is no more reason in the one case
than in the other, for supposing that they did not intend to go back to their
respective homes, as soon as the war was over. Further, the form vaierdacncov
does not differ appreciably in meaning from the ordinary imperfect, evaiov,
which is used for example (1. 681) of the Achaean Hellenes who inhabited
Pelasgic Argos ; nor in the significance of the tense from the /caXevvTo of
1. 684. Had these people then migrated long since from South Thessaly,
and ceased to be called Myrmidons ?

Moreover, even supposing that vaierdaaKov had the meaning which is
suogested, it proves nothing more as to the Thessalian Larisa than it would
prove about any other of the numerous towns of this name. The place-name
Larisa, in fact, is so common in the Aegean, that it is of no practical use as a
landmark. Moreover, so common a name probably had at first a merely
descriptive meaning. What if Adpiaav vat,erdacrKov should be found to have
meant that they ' dwelt in a Burgh ' ? If however it were legitimate to
' count heads ' in such a matter, or to neglect the lateness of our authorities
for all these place-names, the distribution of the name Larisa on the map
would distinctly favour a Hellespontine home for the Homeric Pelasgi as
against a Thessalian ; for a clear majority of the known sites are strewn
down the Anatolian coast, from the Troad southwards, in exactly the same
manner as are, for example, the towns with the name Pedasa, which looks as
if it had the same termination, and occupies the analogous place in the ethno-
logical cycle of the Leleges; and, for that matter, also, as those with the place-
name Magnesia, which has likewise its counterpart on the Thessalian side.

As long as it was thought admissible to regard the Pelasgians as an
< Asiatic people,' 3 any one of these Asiatic towns would have served the
purpose of this passage. And if it were not for the specific mention of Sestos,
it would be tempting to regard these Pelasgians as covering the basin of the
Satnioeis which is not separately mentioned in the Catalogue, though two
heroes are described as coming from thence to the war. 4 But against this

2a Early Age of Greece (Cambridge 1901) i. 4 II. 6. 34, 14. 445. This Larisa might then

p t 172. be identified with a little town of thai name on

on the evidence of II. 10.429; on which the coast about five miles south of Alexandria

see below. Troas.

A HISTORY OF THE PELASGIAX THEORY. 175

identification the following considerations are decisive: (1) it would utterly
dislocate the geographical sequence of the tribe-groups ; (2) this area is
definitely assigned in other Homeric passages to the Leleges/' who (with their
neighbours, the Kilikes) are not mentioned in the Catalogue ; (3) the epithet
ipi/3ob\a/ca is hardly applicable to the mere coast-strip some four miles lone
by two wide, which is all that the Troad Larisa can offer; (4) when the
Pelasgian Hippothoon is killed in 77. 17. 301, it is ttjX' d-rrb Aaplarjs ipifico-
Xa/cos, and the Homeric usage of rrjXe is entirely against its application to
a town only fifteen miles (on a straight road) from Troy, and fully in
sight of it.

It is probable then that the ' deep-soiled ' Larisa of the Pelasgians in the
Catalogue is yet another unidentified site which bore this wide-spread name;
and that it is to be sought, with the Pelasgians of the Catalogue themselves,
on the European side of the Hellespont; not improbably in the low fertile
ground round the head of the Black Gulf, near the site of the later
Lysimachia.

(2)

II. 10. 428-31 : 7rpc<? fikv dXb<; Kape<? Kal Tlaiove? dyKuXoro^oc
Kal AeXeye? Kal Kavtccoves hlot re UeXaayoi,
77/305 Byyu/3/3?;? 8' eXa^ov Avkioc Mucrot t dykpwyoi
Kal t&pvyes nnroSa/AOi Kal M^'o^e? liriroKopvcrTaL

The passage is Dolon's statement of the order in which certain allies of
Priam had been assigned their camping-grounds on either flank — irpb? d\6^.
7rpo? ©u/x/3/3?;? — of Troy-Town. The names are not in geographical order ;
the Karians are separated from the majority of the Asiatic allies, and are
brigaded, so to speak, with Paeonians and Leleges ; and the Pelasgians are
separated both from the Paeonians, and from the Thracians. The latter are
expressly stated in the sequel (1. 433) to have arrived late, and occupied
a separate camp by themselves. The passage would, indeed, have barely
deserved mention, were it not that some modern writers 7 have quoted it i<>
prove that the Pelasgians are an Asiatic people, ignoring not only the whole
tenour of the context, but the further circumstance that whatever conclusions
are drawn from the passage as to the geographical situation of the Pelasgians
must equally apply to that of the Paeonians in the preceding line. Yet no

5 II. 10. 429 (Leleges, without locality, in site.

tli.- camp-passage); II. 20. 92-6 (Leleges Mini : E.g. Busolt, Gr. Qeseh. i.- 165

Trojans inhabit Lyrnessos and Pedasos : //. unter liistorischeu Stammen Kleinasiens. 3 To

21. 86-7 (Leleges live on the Satnioeis 1!.. justify this, he omits the Paeonians from his list :

and Pedasos is their capital). compare p. li»'» ' kleinasiatische P.' Comnare

6 E.g. in the whole Trojan Catalogue only also Holm, Or. Geseh. i. p. 69. 'Sie werden
the Alizones and the Lycians emne Ti)\6d*v : in erwahnt als asiatische Hulfstruppen der Tro-
II. 16. ■!'■'>'■'> Zeus "f Dodona is ti)\66i valwv, janer'; p. 70. ' Nach diesen Stellen (the
i.e. remote from Olympus, or from Phthia, Homeric passages) /n artheilen sind sie ein
Strabo's phrase about the Troad Larisa, Iv ityei Stamm der in Epii ten and Kleina-
re\4tts, is wholly justified when tested on the sien

176 J. L. MY RES

one, so far as I am aware, has ventured to contend that the Paeonians are an
Asiatic people. 8

(3)

Od. 19. 175-7 (describing the peoples of Crete) : e'y puev 'A%aiol,

ev 8' 'KreoKptjTe^ p,eya\i]Tope<;, ev Be K.v8cov€S,
Acopiees re Tpi%di/ce<;, Slot re Y\.e\acr<yoL

Note here, first, that, as the context shows, the object of the poet is to
' add verisimilitude ' to one of Odysseus' many inventions. Any information
which it gives, therefore, may be assumed to have been correct information
for the poet's original audience, as well as for the presumed audience of
Odysseus. The passage therefore describes the populations of Crete as they
appeared at the date of the composition of the poem ; and it is consequently
of the first value as evidence in the present enquiry.

At first sight it is not obvious how a tribe, whom elsewhere Homeric
poets only know as a European people bordering on the Hellespont, should
also have had an abode in Crete. But the context in which the Pelasgians
are introduced seems to supply a clue. Of the other peoples enumerated,
two, the Eteokretes and the Kydones, may probably be assumed to be indi-
genous (in a general sense) ; the former in the east of the island, where
tradition and archaeology alike attest the survival in historic times of a
distinct type of language and culture ; the latter in the west, viral iroha
veiaiov — so to speak — of the mountains of Sphakia. The Achaeans, on the
other hand, may fairly be regarded as a southerly section of the Achaeans of
the Greek mainland ; and these we may accept, on Homeric authority, as
comparatively recent immigrants. 9

There remain the Dorians and the Pelasgians : both — like the
Peloponnesian Achaeans of Herodotus viii. 73 — in an intermediate position,
neither exclusively Cretan, like the Kydones and Eteokretes, nor quite
recent eVr/A-fSe? like the Achaeans of Idomeneus. Anything therefore which
we may infer from this passage as to the Cretan Pelasgians must either be
applicable, provisionally at all events, 9 * to the Cretan Dorians, or there must
be countervailing evidence, of Homeric date, to enable us to differentiate the
two cases. But the latter alternative is out of the question, for Dorians are
not elsewhere mentioned at all in Homeric literature. We are therefore

8 Relying on II. 2. 848-9, 16. 287-8, I make —Zeus.

a present to the adversary of Hilt. 5. 23. 98, " :l 1 1' it were possible to demonstrate that any

where the force majeure of Darius makes them real ethnic or political convulsion occurred in

Asiatic' for a season, as strategical needs do the Aegean after the composition of Iliad ii

here! but before the composition of Odyssey xix, this

'' For the pedigree of Idomeneus see the lines argument would of course be invalidated. This

which immediately follow Od. 19. 178-81, and however is one of those prospective refinements

II. 13. 449-453: it 'goes up to a god,' as in the treatment of these data which, as I have

]|. itaeus would say, in the third generation: explained already, I have felt at liberty to

[domeneus — Deucalion (the Argonaut) — Minos neglect, in the interest of the main argument.

A HISTORY OF THE PELASGIAN THEORY. 177

confined by our present purpose to such inferences only as would hold o- 00 d
equally of Dorians. 10

Now the obvious inference, as to the Pelasgians, is that the Cretan
Pelasgians were so called by the poet because they were known by him to be
a branch of the Hellespontine Pelasgians : they are distinguished from the
old population of the island, and linked with a people whom we have strong
reason for believing to be of more northerly origin ; and geographical
considerations once more confirm the impression that the PelasgTans also
hail from the north. The north wind prevails in the Aegean area°for by fai
the greater part of the year: Homeric sailors at all events were well
acquainted with its behaviour; and Crete, lying as it does like a breakwater
across the mouth of the Aegean, was probably already then the same dreaded
'lee-shore' that it has been ever since, for every boat which goes adrift
south of the Dardanelles. 11 Even on the modern map of Crete, place-names
like ToTroXia, BovXydpovs, StcXafiiSoxvpi, t,cXa{3oTrovXa— perhaps also
Pcoa<To X o>pi and 'Pv<raa-<nriTia ,— are sufficient evidence of what happens ;
and the post-Homeric stories of Phrygian settlements, no less than the
occurrence of Phrygian cults, and of North-Aegean place-names like Aaptaa,
"18a, and the Macedonian II68va( 'Upciirvrva) 1 - and Alov (A/a) go far to
confirm the inference already drawn from the geography.

The mention of Macedonian place-names Recall's us to the question
whether the argument is equally applicable, as it should be, to the Cretan
Dorians of the Odyssey. The non-mention of Dorians on the Homeric
mainland makes it impossible to complete the parallel directly ; but there is
another case of silence in the poems, so significant that it can hardly be due
to chance ; while, if it is not due to chance, it comes very near supplying the
missing link in our reasoning. Of all the coast-line of the Aegean, from
Malea to the coast of Lycia, only one section is unaccounted for in the
Catalogues of Iliad 2. Priam's confederacy ranges, as we have seen, from
Lycia to the Hellespont, and from the Hellespont to the river Axios :
Agamemnon's allies extend from Rhodes and Kos to Peloponnese and the
Western Islands, and thence to Oloosson (Elassona) on the northern frontier
of Thessaly. But of the coast of Macedon itself, from the foot of Olympus to
the mouth of the Axios, there is not a word from the beginning to the end of

10 This point of view was in vogue already in reverse view [Early jig 86)
late antiquity. Andron for example (I'r. 3, ' As it is an tfand far removed from the 1
quoted by Strabo 175) would seem to derive all Greece, it was much less likely to ha
three alike from Thessaly : impelled, no doubt, population mixed by constant advances of other
by the later belief that there were Pelasgians as tribes, such as took place in the histoi
well as Achaeans and Dorians in Thessaly. northern Greece and northern Ita'y.' What 1
robs ».\v olv -Ere^pTjTas k<x\ rols KvSwias avr6- say in the u-xt rests only on my own expei
XOovas u7Top|ot eiKo?, robs 5e \oinovs en-{)\vSas. on that ol the people I hive met
ots e K @(a<ra\ias <pr)a\u eA0(7v "AvSpuv t?";s there, and on the history of A gation
Aa>p/Soj nev irpSrtpov vvv 5e 'E<rTiaia>Ti'5oj \eyo- since Homeric times.

n<vi)s. Hut An lion's guess is neither Homer i - Xa^atuov itthlov at Hierapyti

nor Homeric. A«lp«ra=Gortyna. St. Byz. •. T6prw.

11 Trofessor Ridgeway has taken exactly the

H.S. — VOL. xxvir. v

ITS J. L. MYRES

the epic. Now if the unanimous Hellenic tradition 13 is correct, that the
Dorians of historic times made their immediate entry into Greece in post-
Homeric times, and from the north ; and if, as Herodotus states, in the stage
which immediately preceded that entry they were ' described as a Macedonian
folk,' it would be exactly this strip of coast which would fall first into the
hands of the new-comers, and give them access to the sea. It would be this
strip also, consequently, which would first fall out of the ken of Aegean
political life in the event of invasion from the north. Macedon in fact
was already in the Homeric Age the thin end of the black w r edge of
barbarism, which two generations later was to be driven into the heart of the
Aegean.

In the light of this consideration, the occurrence of a Dorian vanguard
in Homeric Crete becomes not only natural but almost inevitable: as inevitable
in fact, under the geographical conditions, then and now, as the occurrence there
of a vanguard of Pelasgians ; supposing only that the Pelasgians, as the
previous passages have sufficiently suggested, were a people of the north-east
angle of the Aegean, exposed to closely analogous pressure seawards from the
Thraco-Phrygian movement across their Hinterland}^

§ 3 — The Adjective HeXaayi/cos in Homer.

It illustrates well the peculiar methods of criticism which have been
tolerated hitherto, that the two Homeric passages on which the greatest
stress has been laid by commentators on this topic are those in which the
Pelasgians themselves are not expressly named, or stated to exist in the areas
in question ; but where the mere adjective UeXao-yt/cos is used to express
some attribute which in the poet's mind recalled analogous attributes in the
Pelasgians who were known to himself; and where, moreover, it is possible
without going outside the text of the Iliad itself to set up a fair probability
that there were not any Pelasgian inhabitants at the period described in the
poems. The two passages are as follows : —

(4)

//. 2. 681-4 : N?f av rov<; oauot to UeXaayucbv "Apyos ei'atov,
ot t ' AXov ol t ' A\o7ri)v oi re 'Tprjylv evi/xovro,
oi r el-xpv <&6li]v ?}S' 'EXXdda /caXXiyvvai/ca,
Mup/uSoi>€9 8e KaXevvTO KaV'RXXrjves teal ' A-^aiOi.

It will be admitted, I think, that it is a little unfortunate for the supporters

13 Here, as above, p. 173, I am using post- Attica, and in Naxos, which belong, apparently,
Homeric evidence solely to establish a post- to the same immediately post-Homeric period as
Homeric event. those Thracian incursions into Hellespontine

14 If further analogies be desired, they are Asia, which resulted in the establishment of a
supplied by the copious Hellenic tradition of Bithynia. Bui t lie extant evidence for all this
the Thracian settlements in Euboea, in is comparatively late.

A HISTORY OF THE PELASGIAN THEORY. 179

of current ' Pelasgic Theories,' that on the one occasion in the Homeric poems
where the epithet ' Pelasgic' is applied to any locality at all, the poet should
have so rapidly corrected any false impressions which this might convey, by
adding that the people who actually lived there were not called 'Pelasgians '
or anything of the kind, but were in fact specifically ' Achaeans,' and indeed
uniquely 'Hellenes.' Note, moreover, that the UeXacryi/cbv "Apyos of Homer
is a quite different region of Thessaly from that which contains Adpicra.

The difficulty is usually evaded 15 by explaining that though HeXacryiKov

^^709 was held by Hellenes in the time of the poet, or in the time of the

Trojan War, it had once upon a time been inhabited by Pelasgians. and that

possibly descendants of these Pelasgians may have survived as subjects of

Achaean and Hellenic conquerors. All this however is commentator's

inference, not the statement of the Homeric poet; and it will hardly be

contended that a passage like this stands in the same plane of authority

with that in the 'Trojan Catalogue' {II. 2. 843). What it was about the

Thessalian Argos which struck the poet or his audience as ' Pelasgic,' it is

probably too late to determine ; but it may be conjectured that the phrase

may have been suggested by some such remains of early or at all events pi"e-

Achaean fortifications as are so prominent later in Attic legends. No such

connotation however would be possible at all until the Pelasgic name had

ceased to be merely denotative, and had come to be used in just such a general

sense of ' prehistoric ' as would naturally prompt the observation, which follows,

that though the town 16 was of immemorial age, its inJtabitants now were

Achaeans, Hellenes, and Myrmidons, and of quite recent institution there.

And this is all that, for the moment, we are concerned to show. ' Pelasgian "

in fact had already two senses in Homeric Greek: it meant, as a substantive,

certain actual allies of Priam, aud their congeners in Crete: as an adjective

it meant 'prehistoric' — 'that which once was, but most emphatically is not

now.' ( )f course the occurrence of a connotative adjective of this kind is as far

from disproving, as it is from proving, that the facts were as the poet seems to

have believed: 'prehistoric' is not by any means the same as ' unhistoric'

All that I contend for is that if a Pelasgian population of this Argos is ever

assumed to have existed, it shall be on some more convincing data than can

be derived from this pass; _

The other adjectival passage is the phrase in the prayer of Achilles : —

18 E.g. Busolt, Or. Oesch. i.-' K>"> "Das or th is a matter of indifference to the

Bpitheton Pelasgikon setzt jedenfalls voraus, argument. Analog] that in the Cata-

iii Thessalien Pelasger wohnten oder logne, ;<- it stands, a specific toton is intended.

hut hatten: 1 cf. 167 ' so mussten wohl die In any case we must note that to neXaayucbv

ichaitchen und vol hen "Apyos in Home] means a quite different part of

autochthonen Bewohner des Landes gewesen Thessaly from the U(\ao-,iwT:s of Hellanicus

sein.' SeealsoS. Bruck, ','"■ ind that the area "f this

tradiderint (Breslau 1884), p. 5. nf\aoyitvrts is quite differently accounted for

; Whether m ne\atTyiKbv"Apyos means the in the Homeric Catalogue; is the

ranging with Hah-. Alope, and the resl ountry round the Thessalian Larisa.

N 2

180 J. L. MY RES

(5)

11. 1G. 233-5 : ZeO, ava, <\(D&covale, UeXaa-yiKe, TifkoOt vaicov,
A&)8c6v?7? fieSewv Svo-xeifjiipov • dficpi he SeWoi
crol vaiovcr v7ro<f>r}Tai, dvnrTOTrohes, ^a/jiaievvai.

Here we should note, first, that it is not quite clear why Achilles — most
Hellenic of all the Achaeans, according to II. 2. 081 ff. above — should pray in
his deepest need to a Zeus ' of the Pelasgians,' 17 if by this he meant actual
contemporary non-Hellenic inhabitants of Dodona. To assume that Zeus of
Dodona is a local ' Pelasgian ' deity annexed by Achaean conquerors is to beg
the question. Moreover, the more local a deity is, in all ages, the more
restricted is his sphere of influence : for an Achaean at Troy the unqualified
Zeus of the rest of the Iliad, anthropomorphic and TroXvTrXdi'ijros /copra as
the Achaeans themselves, was surer defence than a Gau-gott in Epirus.

Next, the poet of the Catalogue at all events was aware that the actual
inhabitants of Dodona were no more Pelasgians than were those of ' Pelasgic
Argos : ' for 11. 2. 749 expressly describes them as 'Ei/t^ve? (Aenianes) and
Perrhaebians, both of them well-established and wide-spread Thessalian
peoples who persisted into Hellenic times in this region, 1 " and are in no
sense identifiable with Pelasgians. 18 Here therefore, as in South Thessaly^
we have only the name, not the people themselves, in Homeric times ; but
here, fortunately, we have something of a clue, which was wanting wholly in
Thessaly, as to why the Pelasgian name was appropriate to the cult of Zeus
of Dodona.

If there were two points of behaviour on which an Achaean, whether
chieftain, or poet, or audience, was scrupulously careful in daily life, it was in
the use of the bath, and in the choice and arrangement of his bedding. If
there were any two points therefore in which the dancing-dervishes of Dodona
would seem remarkable and repulsive in the eyes of an Achaean, it would be
that they were dvnrroTrohes, yajxatevvac ; and the only possible excuse for
such behaviour in the ministers of a god to whom an Achaean chief could
pray thus as to his own god, would be that this was actually part of the
immemorial observance, and came down from ' prehistoric,' that is to say (as
in Thessaly) from ' Pelasgian ' times.

I admit that at one time I was puzzled by the intrusion, at such a
moment, of details so grotesque and so pedantic ; especially as there was no
evidence either of interpolation in the prayer itself, or of ' late ' tastelessness
in the context ; and consequently no doubt that we have here as genuine
and fervent a prayer as the poet could frame for his hero. But we have only

Busolt, i.- 165, conjectures that Zeus of 17a E.g. Busolt, i. 2 165.

Dodona 'audi rter einheimisehe Gotl der 18 Except of course in so far as Pelasgian can

Pelasgischen Thessaliens war. ' This presupposes be forced to mean the 'Mediterranean Race' of

the existence of a Thessalian Dodona such as modern Italian ethnologists ; and even hen- I

was invented by Unger {Philol. Suppl. Bd. ii. have my doubts whether the populations of

1863, pp. 377 ff.)on the basis oi a note ofSuidas. Pindus would be accepted by ethnologists as in

Cf. Niese, /A-./'. Schiffukatalog. p. 43. any true sense -Mediterranean.'

A BISTORY OF THE PELASGIAN THEORY. 181

to glance at our own Book of Common Prayer to see that the practice
of piling up descriptive phrases in invocation is not confined to Homeric
liturgy; and it does ool need great experience of popular extempore prayer,
to confirm the observation that the descriptive invocations which mean most
to the suppliant are often quite ludicrous to the bystander. What the
function of such descriptive invocations may be is not yet clear. M -
probably tiny are of the nature of a pass-word, intimating to the deity, by
allusion to some intimate quality or mystic rite, that the suppliant is
himself initiate and lit to be heard. But doubtless they serve also to express
and to enhance the suppliant's mental presentment of the recipient of his
prayer ; and also, no doubt, like picturesque abuse, to attract the attention of
a god who, for the moment, peradventure sleepeth.

In this sense then, that he was a god with an ancient and unusual ritual,
Zeus of Dodona may conceivably have been ' Pelasgic;' and certainly not
demonstrably in any other. It is exactly as if a man nowadays should
describe Stonehenge as ' Druidical.' No word is said in the text as to
worship paid by Pelasgians either recent or extinct; and no Pelasgians can
be shown, on Homeric evidence at all events, to have existed in Homeric
times nearer than Crete and the Hellespont.

On the other hand, each of these two adjectival passages, taken literally
and in connexion with Homeric passages solely, does seem to suggest that
adjectivally ' Pelasgian ' meant already not merely 'prehistoric,' but either
positively ' pre-Achaean,' or negatively merely ' non- Achaean ' : that in tact
the correlative — as well as connotative — usage, which predominated in
Hellenic times, was already familiar in the Homeric Age.

6 4. — The Origin of the Connotatm Usage of ' Pelasgian' in Homer.

How did this antithesis between 'Pelasgian' and 'Achaean' arise?
Again a probable answer seems to suggest itself, when ouce we refrain from
contaminating Homeric texts with the later Hellenic commentary. An. _
all their references to earlier times the Homeric poets know no such
universal 'gathering of the clans' as that which rallied to the aid ot
Aleuelaos. The Trojan Expedition then, as Thucydides was aware, was
probably the first exploit — not excepting even the original Achaean Invasion,
which may well have been gradual — which was in the strict sense Panhel
and so the first occasion on which a common designation was required
for the members of the great confederacy, Hence two phenomena : firstly,
a struggle for survival among several generic names, 'Apyetoi, AavaoL, Ayaioly
with a marked predominance of the last named; secondly, the beginnings
— under the literary stress of the compilation of the catalogue — of a
new use of an originally merely tribal name 'K\\//it>. not raerelj
Bynonymous both with the specific Mup/uudovet "//"' with the generic A^aioi
but also as a characterization- word to express connotal vly that dawning
' Hellenism ' which was coming I bhe common bend between chief and

people, as well as between chief and chief. This latter connotati\ - -

182 J. L. MYRES

moreover, comes out more clearly still in the obviously ' coined ' word
UaveWijves in the description of Aias a few lines further on. 1Sa For Aias was
not in the strict sense a ' Hellene' (i.e. a Myrmidon-Achaean) at all.

The Homeric Achaeans, then, were brought to the very brink of
' Hellenism ' by the crisis of the Trojan War ; and in the compilation of the
Catalogue the momentous name came to light. What determined, then, the
choice of a correlative ? In all probability, the same great crisis, and its
sequel.

Thucydides explains the absence of the word fiapfidpovs in Homeric
Greek, Sid to /nrjh' "EWrjvds iron. But the converse also is valid : as soon as
the Hellenic peoples began to feel the need of a common denomination for
themselves, the need arose also for a common word for ' non-Hellenic' The
Homeric poets had however no single generic word for the confederates of
Priam, and the circumstance, that the war was mainly a siege of Troy, made
the name T^we?, and its quasi-synonyms AdpSavoi, Tev/cpoi more nearly ade-
quate than might otherwise have been the case.

Pass on however to the period which immediately followed the war.
Troy-town had fallen; the hegemony of Priam was at an end; extensive
settlements of Achaean ' Hellenes,' as the place-names 19 and the archaeological
evidence show, occurred on the Troad coast ; and the need for a generic name
for the neighbouring tribes recurred with renewed force. Landwards in Asia
Minor, indeed, the old names ' Mysian ' or ' Phrygian ' seem to have remained
in use for the nearest large groups of folk, who were moreover closely akin
to the old Trojans. The Troad itself, with its population always mongrel,
and its varying degrees of Hellenization, easily acquired the descriptive title
of Ato\i'<? — 'patchwork-land.' It was only seawards, therefore, beyond the
Hellespont, that any real difficulty would arise. Now exactly in this
direction the contrast between Greek settler and barbarous native was being
enhanced, during this very period, by that Thracian thrust which we have
already seen to correspond dynamically with the Dorian thrust in the North-
West Aegean ; and with so marked a geographical feature as the Hellespont
between Hellenic Asia and non-Hellenic Europe, it would be only natural to
expect that the correlative to ' Hellene ' — for this corner of the Hellenic world
at all events — would be the name of the dominant or characteristic native
tribe. Now we have already seen that in the Catalogue the dominant folk
in this area between Hebrus and Hellespont are not the Thracians strictly
so-called but the Pelasgi; and it was probably in some such circumstances
as these that the antithesis of "EWrjv and HeXaayos first took rise. 20 From

Iliad 2. 530. ture that the antithesis arose in Thessaly, and

19 The case of Achilleion and Sigeion are was transferred during the Aeolic migration to
typical. The Athenians, in the time of Peri- Aeolis may reasonably be asked first to catch
ander, could claim ovSev naWov AloKevoi psTshv their Thessalian Pelasgians, and then to point
rrjs 'iAiaSos x^P^s fi ov Ka\ a (pirn Kal tomti to the circumstances (if any) other than the
&\\otai, '6 <y o i 'E A A 7) c a> v cruvtTTp^avTo existence of our trans-Hellespontine Pelasgians,
Mev4\ecf> ras 'EAeVrjs apirayds. Hdt. v. 94. which made the transference itself appropriate

20 All who like Busoll <:.0. i.' ; 157 conjee-

A HISTORY OF THE PELASGIAN THEORY. 183

meaning ' pre- Achaean ' in the mother country the name of the Pelasgi
comes now to mean ' pie-Hellenic ' in tins colonial region : but acquires also
now the further connotation of ' barbarous ' which we can trace indeed to
the case of Zeus of Dodona, but which does not otherwise meet us till we
come to Hellenic writers.

We have thus, within Homeric time, a situation in which almost
inevitably the names "RWyv and Ue\a<ry6<; came, in merely descriptive
fashion, to stand for 'civilized ' and 'uncivilized' respectively: so that it was
possible for a Homeric poet to describe either rude non-Achaean fortifications,
or uncouth ritual survivals, as ' Pelasgian,' without intending to convey any
suggestion as to the ethnological status of their originators.

That this interpretation of the evidence is correct is suggested also by
comparison with what happened elsewhere. In the South-East Aegean we
hear little of TleXa cryot ; and in proportion as they recede from view, two
other names Kape? and Ae'X,e7e<? become prominent as generic names for non-
Hellenes. Here, fortunately, in the case of the Carians, the Homeric
evidence is sufficient to show that in Homeric times these folks Avere already
dominant in Caria, and in possession of coast towns; that their speech was
unintelligible to Achaeans ; and that they were philo-Trojan. To this, the
subsequent evidence adds only this : first that the domination of actual
Carians over Caria persisted until the fourth century and later; but,
secondly, that in the interval between Homer and Herodotus, there sprang up
in the South Aegean a great ' Carian Theory ' — in all respects analogous to
the 'Pelasgian Theory' of the North Aegean— in which many ' Carian '-
looking survivals and antiquities, in Crete, in the islands, and even so far
n field as Attica and the Megarid, were construed in the light of the piratical
performances of the real Carians of the vii-vi centuries as evidence of a
wide-spread 'Carian' barbarism in pre-Hellenic times : until, by a strange
inversion of history, it is to a direct ancestor of the Achaean Idomeneus
that the first 'pan-Hellenic' crusade was attributed by the writers of the
fifth century.' 21

An examination of the ancient references to the Leleges leads to a
similar result. An actual people, in Homer, on the Asiatic coast land, they
fade, in Hellenic times, first into the fabled builders of archaic racfioi and
TroXtafxara, then into an ethnologists' label tor pie-historic traits in Messeiiia
and other parts of European Greece. - la

§ 5. — Lemnos, Imbros, and tin Hellespontine Area in Homer.

Before leaving the Homeric data, mention should be made of two groups
of passages, which, though in a sense negative evidence, arc ^>i' some imp
ance when compared with the statements of fifth century win -

One group concerns the population of Lemnos and Imbros in the
Homeric Age. Both islands are mentioned as geographical stepping-stones

- 1 II.lt. i. 171. Time. i. 4. - i 611. Cf. Pal a and Myres, J.ff.S. xvi, 1 1 '

184

J. L. MYRES

between Europe and Asia,' 22 and are quite well known to the poet; but so
far from being occupied by Pelasgians from the adjacent mainland, or by any
allies of Priam at all, they are apparently on the Achaean side. Lemnos in
particular is still the 'city of Thoas,' 23 and ruled by Euneus, son of Jason
and Hyp?ipyle, 24 who had apparently allowed the Achaeans to put in to
Lemnos on their way to Troy,' 25 and traded on provisions at their camp. 26
He also seems to have been of use to them by providing a market for their
prisoners of war, for he bought Lycaon son of Priam from Patroclus with a
Sidonian cup which had belonged to Thoas. 27 Eetion of Imbros carried on
a similar slave trade with Euneus, and in due course bought Lycaon; 28 but,
being a gelvos of the House of Priam, let his purchase escape and go home :
or perhaps this indirect ransom of a princely prisoner was a ' put-up affair '
throughout. In any case there is no trace of a Pelasgian in either island ;
and not only is the Minyan occupation still effective, but a native population
is described, which is twice expressly described as SiWte?. 29 In both passages
they are mentioned in connsxion with Hephaestus ; but they are not stated
to stand in any special relation to him, 30 and they cannot be merely mythical,
for they are dypLocpwvoij 31 and this implies personal experience of them on
the part not merely of the Achaeans but of the poet or his audience. The
2tWte<? do not appear at all in historic times in Lemnos; but we shall
see that a tribe of similar name existed on the neighbouring mainland to the
north in the latter part of the fifth century (p. 205).

The other group of passages concerns the Hellespont, and implies at
the same time a frontier and a tendency to migrate beyond it ; and we shall
be dealing so much with theories of migration in the sequel, that a Homeric
hint of migration in the Hellespontine area must not be overlooked. The
definition of the Thracians in the catalogue as

ocrcrof? 'EWrierTrwro? dydppoos t'l'TO? eepyei 3 '"

clearly suggests that, though the Thracians of Europe were under the over-
lordship of Priam, there existed other Thracians whom the Hellespont had
had not succeeded in confininor. an d who led a more or less nomadic life on
its further or Asiatic bank, like the Galatae of eventual Galatia. That a
Thracian invasion of North-western Asia had already begun in Homeric
times is probable, if only for this reason, that it is almost impossible to say
wl.ere (in the generic sense) Thracian ended and Hirygian began; and it
was only in Priam's youth, we must remember, that the Phrygians themselves
hal pushed up the valley of the Sangarius and fought their great battle

Lemnos, II. 14. 230, 281 ; Imbros, 14.

281.

//. 14. 230.

' //. 7. 467 : 21. 40 12.

II. i. 230

'-"'• //. 7. 467.

-' II. 2:;. 745 7.

-' II. 21. i-Z 1.

S!) II. 1. 554, Od. 8. 294.

:t " In the Iliad they merely pick him up
when he was thrown out of Heaven : in the
Odyssey too (in the mouth of his flighty lady)
they are 'those horrid peop'e' whom he will
iinil when In- t^oes to Lemnos.

:;1 Od. 8. 294.

:i - //. 2. 845.

A HISTORY OF THE PELASGIAN THEORY. 185

with the Amazon-folk. 33 Now if, and when, any such pressure on the European
shores of the Hellespont was in progress, the immediate and inevitable result
would be to squeeze out the Pelasgians of the Catalogue from Europe into
Asia 34 : and exactly this result we shall meet before long.

| G. — Hesiod and the Later Epic.

For the long period which intervenes between Homer and -Herodotus
our sources are unfortunately very few and very fragmentary. They are
sufficient, however, to show that the double usage of the Pelasgian name,
which we have observed already in Homer, was provoking commentary and
speculation ; and they give some idea of the directions in which theorists
were working. The period divides rather sharply into two phases ; an earlier,
in which our authorities are few and mainly epic, and where the allusions
are incidental and explanatory; and a later, in which we are confronted
with a critical and constructive movement, of rapidly increasing originality,
and of a growing complexity and multiplicity both of local traditions and
of schools of enquiry. It will be convenient still, as in the case of Homer, to
keep separate so far as possible the denotative substantival, and the conuotative
adjectival passages.

A.— Actual Pelasgians. — Hesiod (Strabo 327 = fr. 225 Kinkel) is quoted
as saying of somebody,

A(o$(t)i>>]v (f>r)yov re, Tiekaaywv eBpavov, yev

which suggests that he interpreted the Homeric phrase Zev, ava, AcoScorale,
IleXaa-yiKe as if it referred to an actual settlement of Pelasgians at Dodona. 34 *
Now as Homer populates Dodona not with Pelasgians but with Perrhaebi
and Euienes, Hesiod's phrase must imply either dependence on Homeric
tradition for a description of Dodona as it might have been in pre-Achaean
time — in which case the passage becomes evidence not of ethnology but of
current theory — ; or, if it is really descriptive of Dodona as it was in Hesiod's
time (not much before 700 B.C.), it gives us this important addition to our
knowledge, that, as we shall see in the case of Lemnos, the arrival of Hesiod's
Pelasgians at Dodona must be assigned to post-Homeric time. In either case
Hesiod's phrase is no proof that the Pelasgians were autochthonous
Dodona or even existed there in Homer's tin

•. :• 184— IS 3 '"' 1" later times it certainly was not always

• For a very remarkable echo in a late interpreted so. Ephorus for example, who as

writer of this Homeric conception ofa 'Thrako- we shall see was mainly responsible fort]

Phrygian thrust ' see the passage of Apollonius habilitation of Hesiodic views iboul thi Pelas-

of Rhodes in tl p. 222 below. gians, certainly regarded Dodona .1- one of the

Een he is more than followed by Holm, settlements of his Pelasgiau emigrants from

■ ;, . xh. i. 69. 'Hesiod hat dann ausdriick- Arcadia. [Ephoros ap. Strabo 337 tan 5', 5$

lichgesagt, class Dodona der Sitz der Pelasgei <pri<iiv "Z<popos, nc\cury&v TSpvpad &w$£vri • of

war. 1 Surely the must that maj lis 5e Tlekaayol tuv *epl r'i,v 'E\.\<i5o 5wa-

th?t Dodona was a settlement of Pelasgians. arev6vroiv apxawTarot \4-)ovrai.)

186 J. L. MYRES

B. Theoretical Pelasgians. — Asius, who flourished about 700 B.C., is quoted
by Pausanias (8. 1. 4) as follows : —

dvTtOeov he TleXacrybv ev v^rtKo/JbOLcnv opecrcrt
Tata /jbiXaiv 7 dvehwKev, Xva OvrfrSiv yevos eh].

Here for the first time a man ' Pelasgus ' appears as an individual
eponymos ; and also not merely as 'prehistoric,' but as 'primitive,' — the first of
mankind. Pausanias has just stated that ' the Arcadians say that Pelasgus
was the first man who lived in this land;' but an Arcadian origin is
not claimed for Pelasgus in the passage of Asius, and there is no more reason
for holding that Asius believed Pelasgus to have been an Arcadian than
for holding that he made him a Dodonaean or a Thessalian.

Hesiod, similarly, knew of an individual Pelasgus, who was ' auto-
chthonous ' ('Her 10809 he rhv HeXaayov avrb^Qovd cprjaiv eluai). m In this he
agreed with Asius; but he went further when he wrote vleU e^eyevovro
Av/cdoi>o<; dvriOeoLo, ov irate riKje UeXaayoq 37 : for Lycaon is the great
culture hero of Western Arcadia, and the progenitor of a family which
came eventually to include some fifty eponymi of various places and
peoples in Greece.

The contrast presented by these passages from the later epic with the
Homeric evidence is apparent at once. In place either of real people
familiarly known, or vague allusions to a mysterious pa*t we have a clear-
cut theory which represented Pelasgus as the Primeval Man, and consequently
his descendants, the Pelasgians, as representatives of an aboriginal race and
a primitive phase of culture ; and we have also a further stage of theory
in the localization of Pelasgus (and consequently of Pelasgians) in Arcadia,
which is quite foreign to Homer, and marks the first step in a new path
of speculation which we have next to follow out among the writers of the
late sixth century and of the fifth.

| 7. — The Lorjographcrs of the Sixth and Early Fifth Century.

Two distinct movements may be recognized among the Logographers.
On the one hand the method of personification employed by Hesiod and
Asius is applied to other parts of Greece, in which traces of Pelasgians
were admitted. Hecataeus for example 3S makes Pelasgus a king in
Thessaly: for Thessaly, he says, was called Pelasgia enrb HeXao-yov tov
[SaaiXeos. On the other hand, more than one writer, accepting the Hesiodic
theory that Pelasgus was the First Man, were at pains to harmonize this
theory with the claims of other peoples in Greece to be regarded as aboriginal.

Acusilaus, for instance, writing at about the same time as Hecataeus,
seems to have interpolated a personal Pelasgus into the primeval genealogy
of Argos. In this genealogy, Niobe, daughter of Phoroneus, becomes the

88 Apoll. ii. ';. 1, 7 = Hesiod fr. 68 Kiukel. 38 SchoJ. Apoll. Rhod. 4. 266.

37 Str. 221 = fr. Kiukel.

A HISTORY OF THE PELASGIAN THEORY. 187

earliest mortal consort of Zeus and gives birth to Argos. Acusilaus, himself
an Argive, annotates this pedigree like that of the proverbial Welshman
— 'about this time Adam was born' — and inserts Polasgus as a cadet
brother of Argos. The rest of the genealogy is the expanded version
already mentioned of the Hesiodic theory: Pelasgus becomes the father
of Lycaon, 09 fiacnXevaas ' Apicdhwv e/c ttoX\6ji> yvvaiKtav irevrrjKovTa
iralhas iyevvr/ae, including the eponymi of many Arcadian towns, and even
of remoter Greek and non-Greek peoples: Thesprotus, Peucetius, Caucon,
Macedonus, Phthius, Lyeius, and the like. The list ends with Svktivos
or Nuktv/ao?, a 'twilight' personage associated with a Ddmmerung in the
shape of Deucalion's Flood. Meanwhile Lycaon has also a daughter Callisto,
who becomes by Zeus the mother of Areas. Areas, in due course, survives
the Deluge, and becomes the founder of historic Arcadia. How much
of all this was the real sequel to Hesiod's phrase vleis iijeyevovro Avkuovos
avTiOeoto, or how much is later superstructure, is an open question. 39
All that is of importance here is the fact, recorded by Apollodorus, 40 that it
was Acusilaus who engrafted Pelasgus into the Argive pedigree, — ' Phoroneus,
— Niobe — Argos,' — and that this interpolation took place in defiance of the
authority of Hesiod, who had made Pelasgus an avTo^Bwy.

That the expanded genealogy of Lycaon was a systematic attempt to
ascribe 'Pelasgian' ancestry to certain sets of Greeks, especially in the
North-West, is clear from the version ascribed to Pherecydes by Dionysius
01 Hahcarnassus. 41 UeXacryov kcli A^/m^e/p?;? ytverai Av/cdtov .... ovtos
yafiel KvXX/]vr]v pyj'iBa vufj,cp7]v, acp f)<; to opo$ 1) KvWijvi] KaXelrai — this
domiciles Lycaon, as before, in Arcadia — eireiTu rovq ire tovtcov yevvridevras
oie^iwv, tcai riva<; TOTrovq e/cacrroc tovtcov (pKifcrav, OlvcoTpov kcl\ HevxeTiov
p-ifivyjo-fceTai XeyMv mSe : — ktX. Here we have a clear formulation of the
theory of a diaspora of Pelasgian peoples from Arcadia north-westward.
to which system and currency were given later on by Ephorus. And we can
hardly doubt that the goal of this north-westward movement was the
UeXaaycov ehpavov at Dodona, which we have seen reason to believe that
Hesiod had invented out of the Homeric epithet of Dodonaean Zeus.

Hellanicus, a generation later, writing iv 'ApyoXi/cais like the Argive
Acusilaus, makes another and quite different attempt to associate the
lineage of Pelasgus with a genealogy which is essentially Argive. This

3LI This genealogy conies to us as the work of Latesl oi them. Bat Ephorus certainly 1 •

Acusilaus in Apollodorus ii. 1. 1, cf. iii. 8. i.. great mass of genealogical material of earlier

confirmed by Dionysius ol Halicarnassus i. 17. than fourth century dace: genealogical stud, of

'■I ami Tzetzes, Lye. 481. tins elaborate kind is characteristic of tin- later

40 iii. 8. 1. sixth and early fifth century : and in the

41 Fragt. 85 = Dionys. Hal. i. 13. It is a particular case of Lycaon we have evidence that
misfortune that it is not possible to disentangle a metrical gem 1 9 jted which was attri-
with certainty tin' contributions of the three buted to Hesiod. So, rather thai

writers named Pherecydes. Everything that is own view of the matter t" an extreme, I have

attributed to them on this topic i- so situ rated chosen t" discuss tin' statement!

with the ideas of Ephorus sc< ss 1 i 16 below) as if they belonged to the \oyoypi<pos of that

that my own inclination is to assign all to th( nan. - 220 below.

188 J. L. MYRES

theory comes to us in the following form. 1 " 2 Triopas, who stands in the same
eponymous relation to the Dorian hexapolis on the Carian coast, as the
hero Argos does to its Argive metropolis, had three sons, Iasus, Pelasgus, and
Agenor. On the death of Triopas, these ' divided his kingdom.' Pelasgus
took the eastern half, ra irphs ^paalvov irora^ov, and founded Larisa (the
acropolis of Argos City), calling it after the name of his own daughter
(fr. 29). Iasus took the western half, ra 777305 T HA.tf. On the death of
Pelasgus and Iasus, Agenor brought cavalry and conquered the whole
country. This is all to explain three Homeric epithets of Argos : "laaov,
iTnToftoTov, HeXaayi/cov ; and the theory is ascribed to Hellanicus by name.
It presupposes that the name "Apyos was applicable to the whole of the
kingdom of Triopas, which included all Peloponnese ; and so gives us
fifth-century authority for the belief underlying the statement of Apollodorus 43
that the hero Argos on succeeding Phoroneus as king called all Peloponnese
after his own name. Whether Apollodorus' further contribution, when he
puts the hero Argos in place of Triopas, is of earlier date, or is a subsequent
attempt to square the genealogy given by Hellanicus with that given by
Acusilaus, is another question ; and the same observation applies to another
variant given by Eustathius, 44 which puts Phoroneus in place of Triopas :
an even nearer approximation to the theory of Acusilaus.

An obvious motive for these various attempts to interpolate Pelasgus in
genealogies relating to the Peloponnesian Argos has doubtless suggested
itself to the reader by this time. There can in fact be little doubt that
Hellanicus, or Acusilaus, or both, were the victims, if not the perpetrators, of
a simple literary blunder. Hellanicus, it is true, is the first known author
who named TieXaayiooris as one of the Thessalian tetrarchies, 45 and he was as
fully convinced as anyone of the existence of a Pelasgian settlement in
Thessaly down to the time of the ' Coming of the Hellenes ; ' so he cannot be
acquitted of having known that the Homeric HeXaayiicov "Apyos properly
referred to some part of Thessaly. Yet he and his immediate predecessors
are under grave suspicion of having taken that phrase also as referring not to
the Thessalian but to the Peloponnesian Argos ; of having confused both
with that Thessalian Larisa which is neither part of Homer's UeXaaycKov
"Apyos nor the home of Homer's actual YleXaayol; and further of having
combined this non-existent 'Pelasgian Argos' in Peloponnese with the
Pelasgian Arcadia,' which we have seen to be Hesiodic doctrine, and of which
Acusilaus 46 and Hellanicus 47 were both aware.

The actual reduplication of the place-name Larisa, in Thessaly, in
Argolis, and in the country of Homer's actual Pelasgians 48 inevitably in-
creased the confusion, and led to a variety of fresh combinations. Hellanicus 49

Schol. 11. 3.75 = fr. 37. In spite of his « ii. 1. 2. 2.

later date, I class Hellanicus with the other 44 Schol. 11. 3. 75.

genealogists, ami separate him from Herodotus 4,i Harpocr. s.v. rerpapx^ai.

ami Thuuydides, on the ground that all that 4G Apollodorus ii. 1. 1.

we know of his work marks him as a belated '■' Stc}ih. Byz. s.v. 'Apnas.

coiitiuuatoj of the logographic school of history- 43 11. 2. 843.

wiitiny. ia Phoronis fr. 1.

A HISTORY OF THE PELASGIAN THEORY. 189

makes Pelasgus marry Menippe,a daughter of the Peneius, 50 and so localizeshim
in Thessaly, and makes him ancestor of a line of Thessalian kings ; Phrastor,
Amyntor, Teutamidas. Of these the last named is of course suggested by the
ancestor of the leaders of that contingent of Pelasgians in Homer/' 1 who, as we
have seen, are really Hellespontine, and have nothing to do either with
Thessaly or Argolis. Hellanicus again, 52 and also Pherecydes,"' 3 brought
Acrisius the Argive on a visit to the Pelasgians of Thessaly, and so explained
the existence in Thessaly of the Argive place-name Larisa ; and, later,
Staphylus of Naucratis brought Pelasgus himself from Argos to Thessaly to
found this Thessalian Larisa/"' 4 There was however apparently yet another
tradition in the field — perhaps the legend utilized by Staphylus — which put
the foundation of the Thessalian Larisa earlier than the generation of
Acrisius; so yet another step was taken by the defenders of the Acrisius
theory, by duplicating their protege?*

Side by side with all this speculation, one passage from Hecataeus 56
reveals to us an actual population of the Pelasgian name, resident now in
Lemnos, but believed to have once lived in Attica. The passage however is
only preserved to us in abstract ; and we shall be at all events on the safe
side if we postpone consideration of it till we come to discuss the views of
Herodotus, to whom we owe its preservation.

One set of fragments of Hellanicus 57 deals likewise with Lemnos, or
rather, with the ILvTies, its Homeric population. From merely tending the
outcast Hephaestus — and the merest ' Pelasgian' could hardly do less under
the circumstances — they have become his Lemnian craftsmen, 'the first arti-
ficers of metals,' inventors of armour and implements of destruction ; and that
is why they are Sccries, from o-{vea0ac.-' s They are also, by this time,
immigrants from Thrace; for when certain Trojans, who play a part in the
foundation-legend of Chios, landed in Lemnos, rjaav . . . avrodi KaroiKovvTes
©pa/re? rives, ov ttoXXoI avBpwrror iyeyoveicrav Se ^ui^e/vA^e?- tovtovs
eicdXovv ol TrepioiKoi 'Ei'vTias. Their ' Hellenic admixture' we must suppose
to have been due to contact with the Minyans, and perhaps also with
.Agamemnon's Achaeans during the war. For the further hisiory of the
StVTte9 see p. 205 below.

50 Compare the alliance arranged by 'r. 29.

pherecydes between Pelasgus anil Kyllene i'tj<5o ; ' 3 Fr. 26., cf. Schol. Ap, Rh. i. 40, Tzetz.

vvfi<\>t)v (above p. 187); with the result that Lye. 838, Stepb. Byz. s.v. Adpicra.

Pelasgus is localized in Arcadia, fl4 Schol. Ap. Rh. 1.

61 II. 2. 843; Phoronis fr. 26 = Schol. Ap. 55 Apoll. ii. 4. i.

Rh. 4. 1090. It is a further question whether ll.lt. vi. 137.

in the Homeric phrase vlt Svw Aydoto U(\a<ryov " Fr. 1 12

Tet>Tayu/8ao tin- word neXacryov means [ son 58 Philochorus (fr. 6 = Schol. 77. 1. ."94) after

Pelasgus 1 or simply 'the Pelasgian :' and again hi- manner has pounced upon this hit of philo-

whether Tevra/i(Sao means 'son of Teutamidas,' logy and adopted it, hut explains it quite

as Hellanicus thought, or rather 'son .if differently.
Teutamus.'

190 J. L- MYRES

§ 8— The Tragedians.

How popular in the fifth century was this blunder about the HeXao-yi/cov
"Apyos- is well seen from the tragedians. Aeschylus for "example in the
Supplices (11. 1 ff.) makes the king of the Peloponnesian Argos call himself
the son of Palaichthon the earthborn : he is the eponymos of the Pelasgi,
and the lord of a realm which includes everything west of the Strymon,
Paeonia, which he seems to put also west of the Strymon, 59 Perrhaebia,
Pindus and beyond, and the hills of Dodona. It extends, in fact, as far as
the sea, presumably the Adriatic. It also includes all south of this Strymon-
Adriatic line as far as, and including, Peloponnese. Here the genealogical
diagram Tij — Ha\.ai,x0aiv — HeXaayos is clearly an expansion of the Hesiodic
theory of a UeXaayos who is himself avroxOov. The extent of the Pelasgian
kino-dom is no less clearly determined, partly by the desire to include a
' Pelasgian ' Dodona (which had by this time become matter of common
knowledge), and the ' Pelasgic Argos' of Thessaly; partly by an attempt
to claim for the Pelasgian Argos of Peloponnese the hegemony over all
those parts of Greece (including Macedonia) which had come in historic times
under the rule of soi-disant ' Heracleids from Argos.' It is possible also
that the allusion to the Strymon may cover the poet's acquaintance with
the fact, known to Herodotus (i0 that ' actual ' Pelasgians remained extant in
the fifth century within the basin of that river.

In Prometheus, similarly, UeXaayca is used in a context which shows
that the Peloponnesian Argos of Aegisthus and Danaus is meant : —

879 f. UeXaayta ere Several, $i]\vktov(o

"Apet hafxevrwv vvtcTi(f)povp)]T(p dpdaef

and here too Aeschylus is further supported, as we shall see, by the
Herodotean accounts of Argolis as having been Pelasgic at the time of
the coming of the Danaids. 60a

Sophocles m the same way transfers to the Peloponnesian Argos not
merely the associations which belong to the UeXacrytKov "Apyo<? of Thessaly,
but also all that other body of fifth-century doctrine which equated the ' actual
Pelasgians ' of Thessaly, the Thrace ward parts, and Lemnos, with the no less
mysterious Tyrseni.

Fi\ 25G. "\vaye yevvdrop, wal Kpt]vcov

Trarpos 'Qiceavov, fxeya irpecrfievcov

"Apyovs re yvais/'Hpas re 7rdyoc<;

Kal Tvpatjvolcri Tie\aayoi<;. mh

]i an mbe) here L) that Homer's Paeonia is definitely included in 'Thrace,' from which

mil- the river Axius ; (2) that ' Paeonia ' proper is distinct both in Homer, in

igh in the sixth century Paeonia had Herodotus, and even later.

. good way easl ol the Strymon, yet 60 Hdt. 2. 171 ; 7. 94, see §10 below.

;ill th D had been made arao-Toros '" l Hdt. 7. 94 : 2. 171.

in the time of Darius. Hdt. 5. 15. Nordoesit 6ub Dion H. 1. 25. For the Tyrseni see § 17

ever to have recovered its Paeonian below.
icter ; in th< fifth and fourth centuries it

A HISTORY OF THE PELASGIAN THEORY. 191

Euripides contributes little. His regular use of the epithet ' Pelasgiau '
is to denote the Peloponnesian Argos and its population, both Achaean, as in
the Orestes, and Iphigenia in Aulis, 000 and pre-Achaean 3 as in the Phoenissae,
and the Supplices. 606 In Orestes 1247, UeXaayiKor "Apyo? clearly means
Achaean Mycenae. Only in one passage does he distinguish between the
previous IleXacryiMTat and the culture hero Danaus, whose name they are
caused to assume. 01

§ 0. — Herodotus: (a) Iris independence of the Hesiodic School.

With Herodotus we are once more in broad daylight. His allusions to
the Pelasgians are numerous, and his usage of the name, though it varies,
is on the whole intelligible. His work also shows sufficiently clear points of
contact both with recent observation and with contemporary theory to
permit it to be used as a commentary on the more fragmentary utterances of
other fifth-century writers. It has on the other hand the disadvantage
that, thanks to the eclipse which befel the History almost as soon as
it was published, it had surprisingly little influence on the course of
later speculation. But herein there was gain, as well as loss, as we shall see.

Herodotus has, in the first place, no mention of an individual eponymous
UeXaayos ; and no direct contact with the Hesiodic theory at all, except the
bare allusion to the Arcadians as being in the theoretical sense Pelasgians,' ; -
and as being autochthonous in Peloponnese like the Cynurians. 63 But the
Arcadians are in no way specially marked out as aboriginal or Pelasgic ;
and their Cynurian colleagues are never called by him Pelasgian.

This leads us to the positive side of Herodotus' work ; and here once
more we inust distinguish between a writer's accounts of Pelasgians actually
surviving in his own day, or extinguished within living memory, and his
statements of a ' Pelasgian Theory ' of early Greece.

§ 10. — H. /< dolus : (b) actual P< I"*f/ictns as survivals, cl icfly in tht North A< j< an.

Actual Pelasgians, either surviving or recently extinct, are known
to Herodotus in three distinct areas, all on or near the north coast of the
Aegean.

(1) At Placie and Scylace, on the south shore of Propontis, a little east
of Cyzicus, and presumably on the secluded Karadagh plateau. These
Pelasgians still retain their name, and speak a peculiar language which
is not intelligible to their neighbours. 04 Note that these Pelasgians are

Mc Tlfhaayia Or. 060, LA. 1498 ; TltXaoyiKbv * 2 i. 1 16, ii. 171.

"Apyos Or. 1601 ; n(Kaffybi'"A. Or. 602, 1296 ; ,i3 viii. 73 otKtei 8e ri\v Uf\oir6vvr)aov i<?iea

XleXaaybv iSos 'Apytlwv Or. 1247. eTrra • tovtuv ra /J.iv Sio, avrox^oia eoi-ra. Kara

'•"'' U(\acryia Suppl. 368 ; Ut\acrytKuv 'Apyos X^P 7 !" '(SpvTai vvv re nal rb iraAai [ofoeov],

PJloen. 256; neAcurytubv a"rpaTet//.ia 'ApKaSes re Kal Kvvovpioi.

105-6. 57.

,:1 Fr. 227. See p. 221 below.

192 J. L. MYRES

situated immediately across the water from the abode of the Pelasgians
of the Trojan Catalogue ; and exactly in the direction to which the south-
eastward thrust of Thracians, Treres, and Kimmerians in post-Homeric
times had tended to drive the Homeric population of south-eastern Thrace.
Note also that the silence of Homer, not merely as to Pelasgians in Asia,
but as to Thracians in what afterwards became Bithynia, and also the
positive Homeric evideuce as to the non-Pelasgian character of the population
of lemnos raid Imbros, makes a very strong case for assuming that this
Pelasgian occupation of Placie and Scylace results from the same post-
Homeric movement.

A similar raid, by some of these same Pelasgians, reached as far as Attica,
and effected a regular lodgment therefor a time, ol avvocKot iy iv ovr o
' A6)]vaL0i<ri. The approximate date for this raid is given in the parallel
passage in ii. 51: 'A0y]vaioicn yap i]8r) t r\v i k av r a e <? " 'EiWrjva?
t eXeo v er i UeXaayol avvoucoi iyevovro iv rf} ^copy, 6$ev teal r 'E\\t]ve<;
r)pl*avTo vofxiaBrjvai : so that if it is possible to discover at what point
in their history Herodotus thought that the Athenians ' were just beginning,
to count as Hellenes,' it will be possible to assign at all events a relative date
for the time at which these Pelasgians ' came to be fellow-lodgers with them
in their country.' This point however will be best reserved until we come to
the question of the Pelasgians in Attica. 05

(2) In Lemnos and Imbros. These Pelasgians also are post-Homeric
intruders : for they expelled from Lemnos the Miuyans, 60 who are still
in possession there in Homer. 67 Moreover Herodotus fixes the date of the
Minyan migration from Lemnos to Laconia in the same generation as the
Dorian invasion of Peloponnese : for Theras was the brother of the wife
of king Aristodemus. 68 These Pelasgians were still in the islands when
they were annexed by Persia about 505, 09 and were also still in possession
when Miltiades conquered them, not long before 493. 70 It has been argued
from the phrase ere rore viro HeXaoycov ol/ceop,eva<; in v. 26 and from
the omission of Lemnos in the list of extant Pelasgians in i. 57 that
these Pelasgians were extinct when Herodotus was writing ; but he nowhere
states that the Pelasgians were wholly expelled by Miltiades, and in i. 57
he clearly hints at the existence of baa aXXa YleXaayi/ca iovra TroXta-fiara
to ovvofjia fiereftaXe, as though there were people who still talked ' Pelasgic '
and were known to be of Pelasgic origin, but no longer satisfied his
other condition that they should have retained their proper tribal name ;
and this would clearly cover such a case as that of Lemnos under Athenian
rule. 71

See below, §12. 71 Note that a 'Lemnian' who was in the

86 iv. 1 1.'.. Persian service in 480 B.C. (Hdt. viii. 11) counts

67 77. 7. 468, 23. 747, the latter a late as one twv cur jSao-jAel 'E \ A 17 v o: v iovTocv. He

passage. also hears a Greek name, Antidorus. If the

|; ~ iv. 147. Lemnian Pelasgians had not ' chai ged their

i;; ' v. 26. name ' he would presumably have been described

7 " vi. 1 36, as a ' Pelas'dan.'

A HISTORY OF THE PELASGIAN THEORY. ] 93

Now these Pelasgians of Lemnos and Imbros lit', like those of Placie and
Scylace, right in sight of the territory of the Homeric Pelasgians ; and kept up
to the close of the fifth century a piratical connexion with the mainland : for
their Persian administrator had to deal severely with them, rov<; p.ev \nro-
crrpaTtris eVt 1kv6ci<; aiTMOfievos, rov<; Be aiveadai rov \apeiov crrparov airo
Ixvdecov oiricru) aTroKO/ii^ofievovP and from a base in Lemnos they would
only have done this either in south-eastern Thrace, or in course of its transit
over the straits.

The Pelasgians of Lemnos and Imbros were also concerned in early raids
on Attica: for the 'Pelasgians under Hymettus ' in Attica, made Lemnos,
among other places, their retreat : aWa re axfiv X w P^ a Ka ^ %h koi Afjfivov. 73
This connects them directly with the men of Placie and Scylace, whom
we have already seen to be among those oi avvocKoi iyevovTo '\6ijvalota-i.
The raid on Brauron moreover is expressly stated to have been the work of
these same Attic Pelasgians after they had left Attica and settled in Lemnos. 74

(3) In Samotkrace, Herodotus accounts for similar survivals by the
same story. He is illustrating, by the Kafieipcov opyia in Samothrace,
an Attic cult which he believes to be of Pelasgic origin. This would not by
itself prove that there were then, or ever had been. Pelasgians in Samothrace.
But Herodotus goes on to explain, t?]v yap 'Ea/j.odpa/ojv oi'/ceov irporepov
UeXaayol ovroi, oiirep WOijvaioicri a v v o t k o t iyevovro, zeal irapk
Tovrcov 'Ea/xoOp>jiK€<; ra opyia 7rapa\a/x/3avouai. Ua The phrase about the
(tvvoikol is identical, and the present tense of Trapa\ap:[3di>ovcn suggests that
in Herodotus' own time the fountain-head of Cabiric orthodoxy was an
extant Pelasgian community. 7 ''

'-' '•'. 27. Samothracian ritual still used iraKai'av iSiav

vi. 137. The words are part of the citation, SidAeKTov (Diod. 5. 48. 2; cf. Lobeck, Jgl

or summary, of Hecataeus, already mentioned 1109, 1348), and that the cult itself was not

at the end of § 7. then confined to Samothrace. but was observed

74 A closely analogous case is thai of the Isewhere, not merely in Lemnos and Imbros,

Dolopes in Scyros. Originally a mainland and but in the Troad and on the Hellespont (Strabo

inland people, as indeed the rest of them were 472-3), we may reasonably infer that here

still in the time of the Persian Wars (Hdt. iv. also, quite apart from theories, Herodotus is

132, 185), they entered Scyros in post-Homeric dealing with current verifiable observations of

time, and retained their hold on the island North Aegean cults, of the same kind a- I

until they were suppressed by Cimon, as the which he quotes -]■> lineally for the cult of

Lemnians had been by Miltiades. That the Heracles in Thasos. Demetrius of Scepsis, later

Dolopian occupation of Scyros was post- on, had a theory of his own about Samothrace,

Homeric seems to follow from //. 9. 668, which is quite independent of Herodotus, and

where the island is raided and captured by at first sight quite different, but which on

Achilles, and from //. 19. 326 32, Od. 11. 509, closer iuspi ition seems to show that he had

where it still forms part of his dominions. been led by similar data to a conclusion very

The case is here too exactly analogous with similar to thai reached by Herodotus about

that of Lemnos, 11. 14. 230 and Imbros //. these North Aegean Pelasgians. Samothrace, he

14.281. For the further fate of these Scyrinn says (quoted by Strabo 472), was at first

Dolopcs at the hands of the historians, see called MeKirri : the name Zauodpaxr) is

p. 221 below. ad name, and dates from the coming of the

74; » ii. 51. ral>iri, wham to identify with the

From the fact that in Roman times the Curetes. Their cult hi Phrygian.

U.S. — VOL. XXVII. O

194 J. L. MYRES

(4) At Antandrus, on the south-west angle of the Troad, Herodotus lets
fall, unexplained, the epithet rrjv IleXacryt&a™ He does not assert that there
were any Pelasgians resident there in Hellenic times ; but the geographical
position of Antandrus is such as to facilitate settlement there (as at Placie
and Scylace) in the event of Thracian pressure on the country of the Homeric
Pelasgians. That such pressure was felt, and that such settlements were
made, is clear from a fourth-century account of a colony of European Edones,
like those of the Bithynian coast, at Antandrus itself 77 ; and that there was
some non-Hellenic element at Antandrus much earlier than this, is clear from
the phrase AeXeycov ttoXis applied to it by Alcaeus. 78 We have seen already
(p. 183) how closely the ' Lelegian theory,' of which this is one of the most
northerly manifestations, replaces further south the ' Pelasgian theory ' which
prevails in the Hellespont and its neighbourhood. 71 '

The probability that in the fifth century Antandrus was believed to be
not merely non-Hellenic, but positively Pelasgian, in the sense that it held a
population of South-east European origin and post-Homeric arrival, is
increased by the fragment of Hellanicus which is quoted to explain the
proverb Ucrdv?] elfic. ^r^al yap (Hellanicus) avrrjp inrh TleXaaywv
avSpa7roSia0t]vai y zeal irdXiv viro 'Epudpaioov iXevdepooOfjvat.^' Note that
this proverb itself can be traced back as far as Alcaeus, and presumably
the legend likewise, which in that case falls within the class of data
accessible to Herodotus. No dates are given, but the incident must fall
(a) not later than the time of Alcaeus ; (b) hardly, if at all, earlier than
the foundation of Erythrae in the time of the ' Ionic migration : ' for the
poiut of the proverb is that the disasters of Pitane are incessant; so there
can have been no long interval between enslavement and liberation. We
may therefore place the incident in post-Homeric, and probably in very
early Hellenic, times ; and we may class this hint of the presence of raiding
Pelasgians in Aeolis alongside of the other evidence of the kind. 81

(5) Near Crcston, finally, on the mainland between Thrace and Macedon
in the district which lies south-westward of the middle course of the
Strymon, Herodotus alludes to tois vvv en iova-c UeXaaywv rwv virep

The name ~S.afj.nQpa.Ki} in any case looks as if it " s Strabo, 606.

recorded an intrusion from the neighbouring - " 9 The positive statements of Konon Karr.

European mainland, and it is instructive to find 41 and Mela i. 18 that there were Pelasgians at

it suggested that it was a Phrygian cult which Antandrus are only worth noting here as

was intruded, and that its subject was a group evidence of a later revival of .the authority of

of personages, who (like the Hellespontine Herodotus. Mela's version contains an ana-

Pelasgians of Homer) have so exact a counter- chronism and two pieces of thoroughly Graeco-

part in Crete. Roman philology.

7,; vii. 4'2. The phrase clearly denotes some- 80 Fr. 115 b = Zenob. v. 61.

thing peculiar to Antandrus, and not common 81 Pliny N.H. 5. 30. 32 and Steph. Byz. s.v.

to the Greeks of Aeolis. These latter are rbiraAai add Cimmeris to the already long list of ethno-

Ka\(6fxivoi Ue\a<ryol, ws 'EX\i)vwv \6yos (vii. logical epithets of Antandrus.

9f>), but this is Greek theory, not llerodotean Thucydides on the other hand seems tacitly

observation, and is discussed in its proper place to put all this on one side as not-proven,

in §11. when he specifies Antandrus merely as an

77 Aristotle a p. Steph. Byz. s.v. Aeolian colony in viii. 108.

A HISTORY OF THE PELASGIAN THEORY. 195

'Vupcnjvwv Kptja-TGJva tt6\lv oiKeovTwv.^ 1 They spoke a language which,
though different from that of their neighbours, agreed with that of the
Hellespontine Pelasgians at Placie and Scylace.

Much confusion has been wrought in recent commentary on this passage
by the circumstance that Dionysius of Halicarnassus apparently read here
Kporwva for KptjcrTO)va, S3 meaning thereby however not Croton in South Italy,
but Cortona in Umbria, a reading which led him to use the passage as evi-
dence for his own peculiar theory about the origin of the Etruscans. This
reading however has been accepted and defended more than once recently,
and notably by Prof. Eduard Meyer. 84

Those who read Kporoova however may fairly be asked to meet the
following objections : —

(a) Though Herodotus mentions Umbria twice, 85 he uses it merely as a
general geographical expression for northern Italy, and displays no familiarity
either with the country or with its people. It is difficult therefore to believe
that he ventured upon exact philological comparison between the speech of
the people of Cortona and that of the Pelasgians on the Hellespont ; and still
more that there should be truth in it if he did. It is only on the popular a
priori assumption that in a passage of Herodotus an absurdity is more likely
to be the true reading, that the variant commends itself at all ; and it is, in
fact, for the purpose of discrediting Herodotus that the reading KpoTwva is
commonly defended.

(b) On the other hand Herodotus shows himself particularly well
informed about the districts inland of Chalcidiee ; and his descriptions of
lake Prasias and of the road from Paeonia into Macedon have all the look
of eyewitness. 80

(c) His association of Pelasgians with Tvpaijvo! and Kprjarcovaloi is con-
firmed by the statement of Thucydides, 87 who had also special reasons for
acquaintance with this neighbourhood. There are two discrepancies in detail,
(1) that Thucydides is speaking of a mixed population au/j,p,tKTa eOvrj, nearer
the sea-coast, and (2) that he speaks of it as consisting of /3ap/3dpwi/
SiyXcoa-afov. But they do not at all affect the conclusion that Thucydides
either was independently acquainted with the same state of things, of which
Herodotus describes the earlier and more inland counterpart, or was reading
Kpi']or(t)va in the passage of Herodotus which is in question. 88 That such
avpipuKTa edmj should have come into existence nearer the seaboard, is exactly
what we should expect as the result of successive thrusts from one northern
intruder after another. That in the neighbourhood of the Chalcidic colonies

82 i. 57. " 7 iv. 109.

83 i. 29. Sa That In- really knew the country, and that

84 E. Meyer, Forschungen z. alt. Gesch. [Halle there was such a district for even this lias
1891) i. pp. 1 124. been denied latteily— is clear from his reference

i- 94, iv. 19. to rwoTcciia in
86 v. 15-17.

196 J. L. MYRES

the natives should have become bilingual, is again exactly what happens
wherever two cultures meet: the mongrel population just beyond the
Chalcidic ' pale ' learnt Greek for use ' in town,' without forgetting their
own language for communication with their friends in the interior.

(d) There is collateral evidence of community of population between this
neighbourhood behind Chalcidice and those other districts in which an
actual Pelasgian population is best demonstrable. (a) In the case of
Lemnos, the Homeric Sinties, S9 though they do not appear to have survived
there into historic times, have their counterparts in a Thracian tribe, known
to Hellanicus, 90 which has its habitat fixed by Thucydides 91 as lying on
the left flank of Sitalces' inarch from Thrace into Macedon, while Paeonia lay
on his right. It has left its name, moreover, in that Heraclea Sintica,
of which the site is fixed on the right or western bank of the middle
Strymon, a little N. E. of the district of Creston. 92 (/3) In the case of
the Hellespontine area, Herodotus accepts without question a European
origin for the Asiatic Phrygians (who had indeed but recently entered
Asia in Homeric times) and compares them with the Macedonian Bpt/ye?. 93
He also locates Bpvyoi 94 between the ~Xa\rct8ifcbv yevos and the Pieres
(who lay cast of the Strymon) 95 in a list which runs in an order which
is quite intelligible geographically: that is to say, they lay somewhere
between the Strymon and the promontory of Mt. Athos. This all agrees with
the locality indicated with these Bpvyoi, whom he calls ' Thracians,' and who
attacked the army of Mardonius, at a time when its escorting fleet was
destroyed off Mt. Athos. 93 The bisection of the European TSpvyot-Bpiyes
is in turn paralleled by the duplication of the Pieres, some of whom are
east of the Strymon, 97 while others, far west of it, are next neighbours of
the Perrhaebians of Thessaly. 98

Like the Bpvyoi, the Edoni of the lower Strymon, who are also one of
the components of the avp-puK-ra eOvrj of Thucydides, 99 have their Asiatic
counterparts, as we have seen at Antandrus in the fourth century epithet
'HS&W?, where Herodotus had written rrjv UeXaaycSa. The M}^gdones,
also, who for Herodotus 10 ° and Thucydides 101 inhabit a district of Macedon
next west of Creston, north-west of Chalcidice, and east of the Axius, and
survived in Strabo's time as a subdivision of the Edones near lake Prasias, 102
had however by that time almost vanished out of Europe, and were best
studied, like the Pelasgians of Herodotus, on the south shore of Propontis
next east of the Doliones. For Strabo, they are thus immigrants from
Europe, and of the same character as the Phrygians, the Mysians, and the
Doliones themselves. 103 Here, again, no theory is in question : it is simply

89 //. i. .',9k od. 8, 294. '" vi. i.V

; " Fr. 112. 97 vii. 112.

'" ii. 98. ! ' 8 vii. 131. Cf. 177.

,J2 The r P -n<TToovia of Thuc. ii. 99. In Roman " iv. 109.

tim< s there were traces of Zivrui on both banks 10 ° vii. 123-4, 127.

of tin- Strymon : Strabo 331. uu ii. 99-100.

5.3 vii. 73. m» Strabo, tV. 11.

1.4 vii. 185. 103 Stral.o, 566, 575, 736, 747.
'•'■' vii. 112-3.

A HISTORY OF THE PELASGIAN THEORY. 197

a question, how much collateral evidence exists to support an observation of
fact on the part of Herodotus, that a split tribe could inhabit Placie, Scylace,
and the Strymon valley without appreciable damage to its common speech.

(c) The circumstance that Herodotus mentions a Kpijarava ttoXlv
has been criticized in view of Thucydides' statement that his au/x/xiKTa Win)
lived Kara fjuxpa TroXicrfiara. 1 ^ But first, Thucydides' statement refers not
to the people of the district of Creston but to the av/x/xiKTa Wvq of the coast-
land further south ; secondly, it would be difficult to prove, even if it did
refer to Krestonia, that some one or other of these iroklafMara was not called
Kpyja-roov ; thirdly, that there was such a tto^U? in later times is stated posi-
tivelv by Stephanus (s. v.) and an appropriate site for it exists at the
modern settlement of Kilidj.

So far as we have gone, all the Herodotean evidence goes straight
back to the denotative usage in Homer, which makes the Pelasgians a
specific North Aegean people. Only, for Herodotus, instead of being located
on the mainland (with a single offshoot in Crete), they are projected
into the North Aegean islands, and onto the Hellespontine shore of Asia :
exactly as the known stresses of the post-Homeric age would have led
us to guess would be the case. 1 ""' These 'actual' Pelasgians of Herodotus,
moreover, retained still in his time a linguistic character which marks
them as having issued, at an earlier stage still, from a centre of dispersal
sufficiently far back in the Thracian mainland to permit similar projection of one
band of them into the basin of the Strymon ; and so puts their case on all fours
with that of the Herodotean Phrygians. Whether all this observation was
accurately made, is beside the question here, and is not conclusively proved
even by its consistency within itself. All that we are concerned with, here, is
that such observations were not only possible in the time of Herodotus, but
are recorded by him as having been made. It is equally beside the question,
whether they are consistent or not with his general ' Pelasgic Theory,'
which must engage attention next.

| 11. — Herodotus : (c) his general Pelasgic Theory.

If we look now to his connotative use of the name ' Pelasgian,' we shall
find that Herodotus holds a well-defined 'Pelasgic Theory' of the ethnology
of Greece. Once upon a time all that is now called Hellas was called
' Pelasgia ' and was inhabited by Pelasgians. 10 "'" These, in the majority of
cases, have become Hellcnized gradually; and the crucial test of Hellenization is
the change of language from • Pelasgian ' to Hellenic. 1 "•"' , • Herodotus admits
however that it is only by the study of the speech of the ' actual' Pelasgians
discussed in the last section, that any idea can be formed of what ' Pelasgian
speech ' was like.

1114 Hdt. i. 57. Time. iv. 109. '""' i- 56 7, ii. 52-56.

108 For indications of such a movement even " '' i. ."7.

within tin- Homeric age see p. 1S1 above.

198 J. L. MYRES

Of this metamorphosis of theoretical pre-Hellenic Pelasgian into actual
historic Hellene, Herodotus quotes particular instances in several districts of
Greece. Let us take these districts in geographical order from north to
south.

(1) At Dodona, though Herodotus does not definitely assert that
there were ever any Pelasgian residents, he states that the oracle was consulted
by ' the Pelasgians ' in primitive times. 105c He had learned also, apparently on
Dodonaean authority, the theory that in early times ' the Pelasgians' knew no
names for their gods, and only acquired names later, and from abroad. Now
there is nothing in all this which is not obvious ' by inspection ' to any one who
has before him (1) the Homeric phrase about Dodonaean Zeus, (2) the Hesiodic
description of Dodona as YleXaaycov ehpavov, and (3) the Herodotean obser-
vation that ' actual ' Pelasgians talked a language different from Greek. The
reasoning may be formulated as follows. Even without Hesiodic commentary
it might well seem likely to any fifth century Hellene with a ' Pelasgian
Theory,' that the Homeric epithet UeXaayi/ce meant ' god of Pelasgians,'
i.e. of the Pelasgian inhabitants of Dodona. If so, Pelasgians at Dodona,
or their descendants, were calling the god of Dodona ' Zeus.' But
'Zeus' is the Greek name for the god of Dodona; and as the Pelasgian
language is ex liypothcsi different from Greek, the word for 'Zeus' in Pelasgic
must have been different, if there was one. But was there a word for Zeus
in Pelasgic ? Enquiry at Dodona, possibly elsewhere, reveals none ; all the
rSdpftapoi SiyXcoaaoc, who are within hail, call Zeus ' Zeus ' and nothing else.
Yet Achilles addresses Zeus as YleXaayi/ce, ' god of Pelasgians : ' he was
worshipped therefore by them in their unconverted 'Pelasgian' days. In
those days therefore Zeus of Dodona was worshipped as a nameless god,
and is now called Zeus, only because ' Zeus ' is the Greek name for
him. Q.E.I).

(2) In Thcssahj, though Herodotus does not state that there were
Pelasgians there, it is possible that he is assuming their presence when lie
describes the Aeolian Hellenes of north-west Asia Minor as to irdXcu
fcaXeofievot UeXaayoc, &)<? 'EXXr'jvcou A6yo9. 105d The qualifying phrase charac-
terizes this attribution of Pelasgian origin as a matter of current Greek
belief, and as something quite distinct from the ' Pelasgian ' peculiarities of
Antandrus — whatever they were — as has been noted already in § 10 above.
This current Greek belief must mean that these Aeolians represent
either Pelasgians domiciled in Aeolis and Hellenized in situ, or Pelasgians
formerly domiciled in Thessaly, and Hellenized there before their migration
to Asia Minor. In the former alternative, the phrase goes far to explain his
phrase " Avravhpov Trjv YleXaayiSa, but at the same time makes it difficult to
see in what peculiar sense Antandrus was worth calling ' Pelasgian.' In the
latter, Herodotus would seem once more to be putting his own interpretation
on the Homeric phrase to UeXaayiKov "A/5709 which (as we have seen) was, by

"'■""' ii. :>0 52. " ,r ti vii. 95.

A HISTORY OF THE PELASGIAN THEORY. 199

the time of Hellanicus, (1) extended so as to include Thessaly in general, and
(2) confused with Pelasgiotis and with the country round Larisa, with which in
the Homeric Catalogue it is clearly contrasted. 1056 In any case., the phrase of
Herodotus about the Asiatic Aeolians is either fair commentary on the
trans-Hellespontine thrust of Priam's Pelasgians, or else a not-unnatural
interpretation of the phrase to WeXaa-ytKov "Apyos. Here also therefore
we may regard Herodotus as going back to Homeric authority, and as
admitting current Greek belief only so far as it seemed to conform to
Homeric data.

(3) In Attica, Herodotus describes the aboriginal population as a
Pelasgian tribe, the KpavaoL 101 '' Here we have a fresh feature : a Pelasgian
genus subdivided into species with tribal names. There is no Homeric
authority either for Pelasgians or for Cranaans in Attica, nor for any of the
regions which follow, further south ; so that here we are free to regard
Herodotus as summarizing contemporary theory, and perhaps even improving
on it.

These Cranaan Pelasgians of Attica went through, not one, but several
metamorphoses, 107 before they won their way to Hellenism as ' Ionians,' in the
time of Ion, son of Xuthus ; but they had made their first step as early as
the days of Cecrops. Further proof that the Pelasgians of Attica were
Hellenic already at the time of the Ionic migration is given when (in recounting
the origin of the Ionians of Asia Minor, 108 whom Herodotus believed to have
come immediately, though not ultimately, from Attica) the only Pelasgian
admixture which he mentions, in that very mongrel crew, takes the
form, not of Attic but of 'Ap/cdSe? UeXaayoL The quondam Pelasgians of
Attica were therefore no longer Pelasgic when the Ionic colonies were to be
founded.

The passages about Pelasgians in Attica, however, present difficulties of
their own which entitle them to separate discussion later on £ 12). For the
moment it is sufficient to have discovered (1) that 'Pelasgian' for Herodotus
is a genus including tribal species ; (2) that the process of Hellenization was
in some cases capable of analysis, and approximately datable; (3) that the
crucial event in this process was for Herodotus, as for Hellanicus and for
Thucydides, the arrival in the country of some genuine ' son of Hellen.'

(4) In North Peloponncse, fromSicyon westward, there once lived a people
who were Pelasgians generically, with the specific tribal name of AlytaXek. 10 '
These, like the Pelasgian Kpavaoi of Attica, became Hellenized by means of
Ion, son of Xuthus; and then, as fully Hellenized 'Ionians,' migrated into
Attica, and thence again to the Asiatic Ionia.

(5) In the Cycladcs the islanders are, for Herodotus, kcu touto IleXaayLKov

- I>. 17'.' ami 188. i. H<5-

106 v jj_ .,4 l vii. 94.

1 " 7 viii. 11.

200 J. L. MYRES

edvos : but the context 1P -' a does not show whether he means Pelasgian
aborigines, Hellenized in situ, or a branch (like their reputed kinsmen, the
Ionians of Asia Minor) of the Ionized Pelasgians of North Peloponnese. 110

(6) In the Peloponnesian A rgos, Herodotus describes a population, auto-
chthonous and Pelasgian, as receiving from immigrant Danaids the rite
which the Greeks call thesmopkoria. The natives in this case had neither
the name nor the thing. Elsewhere he quotes Danaus (though he was
not ' a son of Hellen') side by side with Xuthus, as one of those whose coining
marked the crisis before which the people of all North Peloponnese iicakeovTo
Ue\a<ryol AlyiaXels. Another point of theory emerges here. Hellenism in
the sense of the operation of a ' son of Hellen ' is not the only form of en-
lightenment. Danaus from Egypt can ' Hellenize ' in a generic sense : at all
event.- his arrival troubles the Pelasgian waters with the movement of
a new spirit. Have we perhaps here a reminiscence of the phase, which
we conjectured earlier, 111 when Danaus competed with Hellen for eponymous
rank in Greece \

Meanwhile it is clear that though Herodotus may perhaps have shared
with his contemporaries the current misconception as to the Pelasgian
claims of the Peloponnesian Argos, there is no evidence that for him
this district stood in an}" such special relation to Pelasgian antiquity as had
been assumed recently by the genealogists.

(7) In Arcadia there were 'Ap/caSe? UeXaayoi, — again apparently a
specific sub-division of a Pelasgian genus, — who took part in the colonization
of Ionia. 11 ' 2 The Arcadians also were regarded by Herodotus as the sole
survivors 113 of the aboriginal population of Peloponnese ; and this aboriginal
population was apparently continuous with that of ' Pelasgian ' Argos. On
the other hand, in his formal survey of Peloponnesian ethnology, 114 though
he classes the Cynurians with the Arcadians as autochthonous, he omits
to call either of them Pelasgians. We cannot say therefore that there is in
Herodotus any preferential treatment of Arcadia as a source, or habitation, of
Pelasgians.

(8) In Cynuria the same remark applies. Though autochthonous,
the Cynurians are not called Pelasgians : their pedigree is taken only
so far back as to describe them as ' apparently Ionians,' who have however
since ' become thoroughly Dorized.' n " Here we get a fresh point of
Herodotean theory. Hellenism, like Pelasgism, is a genus which includes
diverse species. ' Ionian ' Hellenism is one type, ' Dorian ' Hellenism is

As in the case of Asiatic Aeolis (2) above. evidence, this tempting guess remains

110 It there were any early evidence for the unverifiable.
legends ofThraciansinNaxos and other Cycladic 1U p. 181, above.

islands, it would be tempting to regard this n - i. 46.

ascription of 'Pelasgian' origin as a hint of 11:: ii. 171.

raids of Hellespontine Pelasgians like thos 1U viii. 7.S.

which we have detected already as' far afield as "•"' viii. 73.

Crete and Attica. But in default of such early

A HISTORY OP THE PELASGIAN THEORY. 201

another; and it is possible for avr6)(6ove<; to undergo conversion, not merely
from outer darkness to any one of these types of enlightenment, but from
any one sect to another. The latter process, like the former, is a long
one : eKheScopievprai Be, he can say of the 'Ionian' Cynurians, vtto 'Apyeioov
ap%6fievoi Kal too y^povov.

§ 12. — Herodotus: {d) the Pelasgians in Attica.

Between the statements of what I have called 'Pelasgian theory'
in Herodotus, and his accounts of Pelasgian tribes either actual, or only
recently extinct, lies one group of passages which has caused some perplexity,
but seems to me susceptible of simple and instructive explanation. The
people of Attica, as we have seen in § 11, are for Herodotus autochthonous
Pelasgians, who ' became Ionian ' and so entered the Hellenic family,
in the days of Ion son of Xuthos. 110 On the other hand, just at this very
phase 'A6)]vaioicrc ?;8>; r^vcKaura i<;'R\\riva<; rt-Xeovcri, UeXacryol ctvvoikol
iyevovTo ev rfj x^PV '■ II7 an d by the side of these Pelasgian ' country
cousins,' the autochthonous Attic Pelasgians really seemed quite civilized,
odev Kal "E\X)^e? ijp^avro vofuadfjvai. Of these intrusive and relatively
recent Pelasgians, Herodotus gives further particulars, partly on the authority
of Hecataeus, partly from local Attic tradition.

(1) He quotes Hecataeus to the effect that it was these Pelasgians who
built for the Athenians the wall round the Acropolis. This reveals, as one
element in the story, an aetiological myth about the so-called Pelasgic Wall,
which was still defensible in the days of the Peisistratidae and may be iden-
tified with some certainty as that Mycenaean fortress-wall of which remnants
are still to be seen. Of the open space below this wall, which Thucydides
knows as to Ue\aayuc6i>, neither Herodotus nor Hecataeus has anything
to say.

(2) He quotes Hecataeus further to the effect that these Pelasgian
wall-builders were allowed to settle in the country vtto top TpLtjaaov, that is,
as the story shows, between Hymettus, the Ilissus, and the Saronic gulf.
This repeats (what we already know) that these Pelasgians are not auto-
chthonous in Attica, but recent immigrants ; and it takes this belief as far
back as Hecataeus.

(3) Eventually those Pelasgians misbehaved, and were expelled ; and
went and occupied aWa re . . . ^wp/a Kal Sr/ Ka\ Arjfivov. 118 This also
comes from Hecataeus, and consequently goes back to a contemporary of
the conquest of actual Pelasgians in Lemnos by Otanes, between 510 and
500 B.C. ; and also of their conquest by Miltiades, which belongs to the
same generation. 119

116 viii. H. I" vi. 137.

117 ii. 51. »» v. 26. (Otanes] : ri. 136-140 [Mittiades).

202 J. L. MYRES

(4) Local Attic tradition added this, 1 - that after settling in Lemnos,
some of these Pelasgians returned and raided Brauron on the east coast
of Attica ; and that they did this €v i^e7ncrrup,€vot tcl? 'A$i]va(cov oprds,
presumably therefore within the lifetime of those who had been themselves
expelled from Attica.

(5) The Pelasgian occupation of Lemnos is assigned by Herodotus to an
ascertainable date. In Homer, as we have seen, 1 - 1 the Minyans have not yet
been expelled from the island. They were however expelled, according to
Herodotus, 1 " 22 in the third generation of the Argonautic occupation; that is, in the
generation after the Trojan war, for Euneus, who is king of Lemnos in Homer,
is the son of Jason, who occupied the island. But here there is a slight hitch
in the story. The local Attic tradition, as we have seen, attributed the raid
on Brauron to Pelasgians who were ovtol Arjpivov Tore vep,6/u.evoi ; in which
case the raid was subsequent to the occupation of Lemnos. But in telling
the story of the Minyans, Herodotus says that they were expelled by Pelas-
gians ro)v e/c ¥>pavpwvo$ Xrjio-a/JLevuiv ra? 'Adijvatwv jwaiKa^. It is possible
that he merely adds this detail for the sake of identification, and without
intending to say that they had already raided Brauron ; but at first sight
it certainly looks as if he meant to put the raid before not after the occupation.
And there is this further evidence in the same direction. The rest of the
story of the Minyans dates their eventual arrival in Laconia within the
generation (jov 8e avrcv tovtov xpovov) 1 ' 13 of Theras, great-great-grand-
son of Polyneices of Thebes, and brother-in-law of Aristodemus, about
the time of whose death the Dorians conquered Laconia ; and this entry
was fully two generations after the Trojan War. Either therefore we must
allow the best part of a generation for the 'exodus' or ' iwto? ' of the
Minyans, or else there must be a misfit of one generation in the chronology ;
and in the latter event it may well be the reason why there is ambiguity as
to sequence of the occupation of Lemnos and the raid on Brauron. But there
is no serious inconsistency; and though the whole story comes to us from two,
or more probably three, independent authorities, — Hecataeus, local Attic,
and perhaps local Laconian tradition, — we are in a position now to fit it all
together as a single series of events, of brief duration and approximately
ascertainable date ; for it falls in any event within a generation of the Dorian
invasion of Peloponnese.

According to Herodotus therefore — and I do not claim at present any
earlier authority for this version, — once upon a time there were Pelasgians
in Attica, in the same sense as there were Pelasgians everywhere in Greece in
pr. '-Hellenic days. Just as these Attic Pelasgians were beginning to 'count
as Hellenes,' in the days of Ion son of Xuthus, 124 Attica was invaded by quite
a different sort of Pelasgians, of the Hellespontine variety who survived at
Placie, Scylace, Lemnos, Imbros, and Samothrace. His repeated phrase

. 138. '- :; iv. 147.

121 II. 1 1. 230 and §2 above. I - 1 viii 44.

122 iv. 145.

A HISTORY OF THE PELASGIAN THEORY. 203

o'l avvoiKoi eyevovro *A6))i>atoio-t can hardly mean anything else than that
this Hellespontine type of Pelasgians is the source of the invaders of Attica ;
though no doubt, as in the case of Lemnos, Pelasgians ejected from Attica
retreated in a direction where there were settlements of their own countrymen.
As we have fifth century authority for the contemporary existence of Ion son
of Xuthus and of Theseus, and as Theseus was himself an Argonaut, we can
assign the invasion of Attica by Hellespontine Pelasgians to the generation of
the Argonauts approximately ; and as their expulsion from Attica occurred
not earlier than the first generation after the Trojan War (i.e. the third of the
Argonautic occupation of Lemnos) and not later than the second, we can give
to it a duration of about three generations, and an approximate date within the
fifty years which preceded the Dorian invasion. Within these fifty years falls
the raid on Brauron, a second attempt of Hellespontine Pelasgians to get a
footing in Attica ; but whether of fresh Pelasgians from Hellespont, or of
ex-Attic Pelasgians from Lemnos, remains in doubt. Within these two post-
Trojan generations fall also the Pelasgian occupation of Placie and Scylace
(in a neighbourhood which, for the Catalogue, is not Pelasgian) and probably
also the settlements in Imbros, Samothrace, and the like : for Imbros also has
no Pelasgians in Homer, though it had already, as we have seen, a Sintian
population, which to fifth century eyes 125 must have seemed to be of mainland
origin.

Now we have seen already that the department in which Herodotus
seems to have struck out a new line of Pelasgian enquiry is in the collection
of evidence of the survival of actual Pelasgians in the North Aegean, round
the fringe, so to speak, of the Homeric Pelasgians of king Priam ; and I do
not think that we are unduly straining the sense of the passages which deal
with the Pelasgian invaders of Attica, if we regard these also as a contribution
to the same enquiry.

That Herodotus regarded some part of the population of the promontory
of Attica as still of non-Attic origin, is suggested further by the terms of his
comparison between Attica and Scythia. In this comparison, when once
allowance has been made for the geographical conceptions of the fifth century,
all the other features quoted are markedly apposite; and when he goes on to say
Kal Trapa7r\ij(ria Tavrr) Ka\ oi Tavpoi vepovraL -/}■? 'S.kvOikijs, &)? el rij<; 'ArTCicr}?
aWo hdvos ical /xrj 'A0i]valoi ve/xotaro tov yovvbv rbv —oviuaKov, /ctX., it is
difficult not to believe that, although he does not mention them, he has the
vision of non- Attic Pelasgians in his mind. It may indeed have been common
knowledge in his time that these predatory Pelasgians had had a footing about
Sunium, as well as ' under Hymettus.'

The Herodotean phase of the 'Pelasgian Theory 5 may therefore be
summarized as follows. The logographers have done their work : they have
multiplied Pelasgian origins to such an extent that it is possible already to
generalize. All Greece, in fact, was ' Pelasgian ' once, and the large majority

I "' Time. '1. ! |s . Herodotus' in the G rnal, viiL

126 See my paper 'On the Maps used by 1896, pp. •

204 J. L. MYRES

of actual Hellenes are by descent Pelasgians, Hellenized. But 'Pelasgian' has
now ceased to be a race-name, and means the pie-Hellenic phase of divers tribes
whose proper names are known. There is even the beginning of a tentative
and unformulated theory of how Hellenizatiun is effected. In the light of
this Pelasgian generalization, and of the new ' Hellenic Theory ' which is its
corollary, the special claims of Dodona, Thessaly, Arcadia, and the Pelo-
ponnesian Argos, are seen to fade away. Attica, on the other hand, begins
to rise to new prominence in the story ; 127 due partly to the recent active
contact between Peisistratid Attica and the 'actual' Pelasgians of Lemnos ;
partly to the contemporary desire to find some historical explanation of the
rapid rise and peculiar characteristics of the Attic State since Cleisthenes ; 128
but partly also to the increased importance which the fifth century is coming
to attribute to the evidence of cultural survivals, in comparison with that of
place-names or of literary or oral tradition. Philology and Genealogy, in
fact, are rapidly giving place to Anthropology as the instrument of historical
research. And anthropology while it has nothing to say of Thessaly, and can
prove only foreign influences in Arcadia, has already detected numerous cases
of survival in the neighbourhood of the Homeric Pelasgians on the Hellespont,
together with a true cause for their actual distribution. And when we come
next to consider the attitude of Thucydides to the question, we shall find the
same tendency predominant.

§ 13. — Thucydides.

From Thucydides, with his extraordinary concentration upon those
aspects of history which he regards as his proper concern, we should not
naturally expect much light on questions of ethnography. It is therefore the
more instructive to find that on the rare occasions on which he does digress
into such matters, his knowledge and his beliefs not only agree in general
with normal fifth century views as we find them in Herodotus, but also,
where they diverge from these at all, do so in directions which foreshadow
exactly the principal new departures which are to characterize the speculations
of the fourth century. In this, in fact, as in much else, Thucydides stands
just at the parting of the ways.

A. First, as to actual Pelasgians. Thucydides gives an account of the
natives of Mount Athos, 120 the substance of which we have already noted in
discussing the evidence of Herodotus. Its main points are as follows: —

-" In Attica also, alone, do we find the I have collected some evidence for the view

'theoretical' and the 'actual' or rather the that a .similar demand of the Fericlean Age to

orical Pelasgian side by side in the know 5T %v alrfyv there was a Delian League,

same context, contrasted as Hellenizable Attic was producing very similar effects in a reasoned

aborigines againsl savage Hellespontinu retrospect of Mediterranean sea-power.
intruders. '-'■' iv. 109, seep. 196 above.

In an earlier essay {J. U.S. xxvii. 81 II'.)

A BISTORY OF THE PELASGIAN THEORY. 205

(1) The promontory, as its physical position would suggest, was a
gentium. Its population included wail's from all the principal native storks
of the adjacent mainland: Edones from beyond the Strymon, Bisaltae and
men of Creston from between Strymon and Axius, and Pelasgians t&v /cat
Arjuvov 7roTe Kal ' ABtjvcK; Tvpai]vcbv oltcrjo-avTOiv : a phrase which sums up
all the main features of tin- Herodotean diagnosis of ' actual Pelasgians' in
the North Aegean, except that he makes no mention of their kinsmen on
the Hellespont,

(2) The mention of Creston, as we have seen already, is important
confirmation of the manuscript reading of Herodotus i. 57.

(3) These people are avfifiiKra eOvq fiapfidpwv hiyXdiaawv. If fiap-
/3<ipa)v, one of their languages was non-Hellenic. What their ' second
language ' was, is not stated ; but we may fairly infer that it was Greek : for
though ' barbarian,' these people are in the heart of Chalcidice ; and, as
Thucydides says, Kal rt Kal XuXklZlkov hi /3pa^v. For the rest, they
presumably retained each his own native dialect ; that is to say, the
Pelasgians among them still talked Pelasgic, exactly as Herodotus says of
their namesakes up-country.

(4) Though Herodotus does not actually say that Pelasgians of the
district of Creston were among the colonizers of Attica, he does say so
of the Hellespontine Pelasgians; and these he connects with those of Creston
by the significant tie of a common dialect. In Thucydides, either wo
have additional evidence for this identification, coming from a fresh quarter,
and from a writer who had peculiar opportunities for enquiring locally ;
or we have a fresh inference from the data supplied by Herodotus, in
which case we must infer that these data were accepted by Thucydides
as trustworthy so far as they went. The importance of this latter point
is obvious, in view of the captious attitude which Thucydides usually
adopts in dealing with his predecessors; and, no less, in view of modern
attempts to show that Herodotus in this passage is describing CortOna
in Italy !

(5) Thucydides has also one small piece of confirmatory evidence in regard
to the general view of North Aegean ethnology, the history of which we are
tracing. It is he who is our earliest authority for the existence of those
llurot in Thrace, whom we have already had occasion to compare 1S0 with the
Homeric Zivnes of Lemnos. Here also the strength of the evidence lies
in Thucydides' special facilities for exact knowledge of to. eVl ®patcr)<i;
and, with this admitted, the significance of the reference, in Herodotus vii.
223, to a town ItVSosnear Therma, becomes obvious at once.

B. The Pehtsrjiun Theory of early Greece, which is found in Thucyd
presupposes that of Herodotus, but differs from it in details, which all
mark advances in historical method. 1,1

ii. 98, cf. p. lSt above. lsl »■ 3 -

206 J. L. MYRES

(1) Thucydides recognizes that the ' theoretical ' Pelasgians have their
name from some single tribe, which really was called Pelasgian, but did
not constitute the whole or even the majority of the pre-Hellenic population
of Greece ; kOvrj he aXXa re /cal to UeXao-yi/cbv eVi irXelarov cup' iavTWv
ttjv eTrcovv/JLiav irape-^eo-Qai. We may fairly infer from this that since
the time of Herodotus a still wider induction has been attempted, based
upon data derived from those parts of the Greek world where the pre-
Hellenic population had been previously labelled Carian, Lelegian, Caucon,
and the like, as well as from those where it had been labelled Pelasgic.

(2) The Pelasgic name has consequently acquired for Thucydides a
definite generic and connotative value, which is distinct from its specific
and denotative use as in iv. 109. For the first time, that is, a Greek
historical writer is using a Pelasgian hypothesis consciously, with the
knowledge that it is a hypothesis, and not a summary of observed or
reported facts.

(3) Whereas Herodotus rests content with a view of the process of
Hellenization which is expressed intransitively 132 and assumes a kind of
spontaneous generation 133 — ' spec's I growed ' as Topsy said — Thucydides
is conscious that to 'E\Xi)vik6v has arisen by actual contact of 'Pelasgian ' non-
Hellenes with a body, however small, of genuine and actual "EXA/^e? who
had the higher culture, and so were ' of use ' to their neighbours. Of course
the discovery that Hellenism spreads by contagion only puts the problem one
stage further back : for the obvious question is now, how to account
for the real Hellenes. But it is a clear advance to have formulated the
view that culture does thus come by contagion, /cad' e/cacn-of? fiev ?;S?;
ttj ofiiXla /xdXXov KaXeladac "EiXXijpas ; that it is quality which tells, not
quantity; that 'a little leaven' may work 'until the whole is leavened';
and that, like the Pelasgians, the Hellenes have come to have their name
used in a connotative as well as in a denotative sense ; of which indeed
we have seen the vague beginnings already, in Herodotus' use of Danaus
side by side with Xuthus. But we find no express formulation of it till
Thucydides puts ' Danaans,' ' Argives,' and 'Achaeans 'as equivalent Homeric
names for those 'men of Phthia' olirep kcu irpwToi "EiXXrjves rjaav.

(4-) Thucydides makes no doubt that the real Hellenes first became
appreciable in Phthiotis. What then becomes, for him, of the view which we
have seen growing up in post-Homeric times that to UeXao-yi/cbv "Apyo?
was a hotbed of Pelasgi ? Surely here if anywhere the Hellenic ' leaven '
must have ' worked ' early and effectually. The process of Hellenization
was gradual and lengthy, as he admits; ov fievToc ttoXXov ye "fcpbvov
rjhvvaTo Ka\ a-rraaiv eKviK?)o-ai : but missionary enterprise, like charity, surely
begins at home.

1 ■- to o0vo/j.a /AeTffiaAf . . . tt)c •/Xwooav /act- ' ; Though even Herodotus associates in

ef4.a8e i. 57 : rb 'EAA. anoffx'ffdii' fiivrot curb rov some cases the crisis of Hellenization with a

TleAacryiKou i. 58 : aireKpldr) e/c iraAatrtpov rov ' child of Helleil ' Slich as XuthllS.
^at>$d.pov tdveos rb 'EAA. i. 60.

A EISTORY OF THE PELASGIAN THEORY. 207

§ 14. — The Comparative Method, in Thucydides and in tTu Early Fourth

Century.

In a neighbouring passage 1114 Thucydides formulates — also, I believe
for the first time in literature — the 'comparative method' of ethnological
enquiry. Ceteris paribus, he argues, it is permissible to infer from the
present state of a backward people to a previous state of an advanced people.
It is possible therefore to plot out, in a series, all known varieties of ' Hellene,'
from the most cultured to the least ; and as Hellenism, for Thucydides,
stands for the highest form of culture, the most cultured will be the most
truly Hellenic, and the least cultured will show the most purely Pelasgian
survival.

We, who have passed more recently through a similar phase of method,
know only too well the corollary which a looser logic may allow to be drawn
from such a series. Granting, as everyone did grant, including Thucydides, that
early Greece had been the scene of intense 'distress of nations' and long-
continued [xeTavaardaeis, it was only too easy to confuse cultural with
geographical advance; and to argue (as the students of "Aryan languages' argued
repeatedly in the last century) as if those Greeks who had ' progressed least'
in culture had therefore 'advanced least' from a geographical focus of
dispersion. Now if the zero of advancement is the ' Pelasgian ' stage of
culture, the starting point of Greek /jLeTapucrraai^ ought to be the
'Pelasgian Home,' to adapt a familiar expression. Thus all that was
necessary, in order to discover inductively the Pelasgian Home, was to arrange
all Greeks in their cultural order, and see whereabouts on the map the most
backward of them were to be found. 1343

Now in the early fourth century, the answer to this question was easy ;
and it was threefold. (1) Only one people in nearer Greece (apart from
districts like Messenia and Thessaly which had neither shaken off nor
absorbed their ' conquerors ' since the late ^Tavaardae^) had failed to adopt
in full that 7ro\i?-system which alone — so Thucydides, and Euripides, and
Plato thought — could produce or sustain Hellenic Man : only one people in
all Peloponnese answered to Thucydides' description of his 'actual' Pelasgians,
Kara 8e [xiKpa 7ro\ia-/j.aTa oiKovaiv : 135 only one area had so far ignored the
trend of Hellenism as to permit its sons, in that clash of principles
which was d^ioXoycoraTov rwv irplv, to fight for either side indifferently : 1::0
and that was Arcadia and the Arcadians.

(2) On abroad review of the culture of Greece, the full Hellenism of
Athens and the Ionian 'colonies,' of Corinth, of Argos, and of Delphi, might be

m i. 6. PovAofxtvoi to ir\et<TTa twv irepl avrovs. To

134:1 A very .similar fallacy confounds advance this frame of mind belong.-- also the Enhoran

in culture with progress in time. Ephorus is a theory of the longevity of 'primitive' men

conspicuous instance (Fr. 6 = I)iod. Sic. 1. 9) (Fr. 24 = Plin. N.H. 7. 18) 'Ephorus

irepl irpwTuv Si twv # a p $ d p oi v 5it£i/xev, ovk Art ' anuos vixisse.'

a p x<* i o t * p ov s avrovs ijyovfievoi rwv 'E \ A t)- u ' iv. 109.

v u v Kaddnep "E(popos ttpi}Kei', ak\a irpo%Lt\dzlv 136 vii

208 J. L. MYRES

figured as fading away gradually north- westward, into a region where, first, as
Thucydides well knew, 7r6\et<; gave place to a life Kara /cco/ia? aTeix^crTov; in
Aetolia, and where even hoplite armour was unknown, as in Locris; 137 where,
next, Hellenic speech became blundered and confused, so that Demosthenes'
army had need of interpreters, 138 and he could trust to his Messenians being
taken for Peloponnesians by their accent ; where, farther afield, Peloponnesian
troops feared massacre inro twv /3ap/3dpcov kcu iy^Oiarwv 'AfK^iXo^cov ; 139
and where, behind all, and on the extreme edge of the Hellenic world, lay
the rude ritual, and the immemorial age, of the oracle of Zeus at Dodona.

(3) On a still broader view of the civilized world, the march of culture
was still more clearly seen to be westward. Danaus, 140 Pelops, 141 and
Cadmus 142 had brought 'light from the East' to Hellenic lands; ' Hellen
and his sons' 143 had spread their own light not only to Dodona, but also to
Magna Graecia and to Sicily. But round these western outposts also lay a
penumbra of barbarism, and beyond, a great expanse of peoples who, like the
'theoretical' Pelasgians of Greece, 144 €7rayop,evcov avrov? eV &><£e\/a,
Kad' eKticrTovs [xev ijSij rfj 6/xi\ca /udWov were becoming severally confronted
with Hellenic culture, whose receptivity of things Hellenic was remarkable,
whose cults and legends bore strong resemblance to the ruder phases of
Hellenic religion; who continued to practise a 'Lesbian rule' in their
architecture, which recalled the primaeval citadels and terrace-walls — the
UeXaayitcd relxv — of old Greece ; and whose coasts were still infested by
the lawless pirates whose name in the Aegean was already thrice associated
with the Pelasgian, 145 and who had made the Lower Sea ' Tyrrhenian ' for
good and all. Italy and the West were rapidly being involved in an enlarged
Pelasgian Theory. 14 "'-'

What precedes is, I believe, legitimate inference as to the probable
course of speculation, from the position taken up by Herodotus, along the lines
which are suggested by the indications of advancing method in Thucydides ;
and it accords with the actual extensions which Pelasgian theory received
during the next generation. A crucial instance will make the situation clearer.

i:J7 iii. 94, 96, cf. 112 (Amphilochia). elaiv.

IiS The Ophiones and Eurytanes were ayvw- un Hdt. ii. 9S, 171, 182, vii. 94.

aroraroi yAuxrcrav, i<a\ LO/j.o(pdyoi elai'v, ws \4ye- 141 Hdt. vii. 8, 11.

rat. Time. iii. 94. The Messenians lie de- u - Hdt. ii. 44-49, iv. 147, v. 57-8.

Clibes as Awpida re yAwaaav livras Kal rots u:; Hdt. i. 56, 60. Thiie. i. 3.

■7rpo<pv\a£i Trln-Tiv irapexo/JLii'ovs. iii. 112. m Thuc. i. 3.

: " iii. 112. Of these same Amphilochians Ui Hdt. i. 57. Soph. Fr. 256. Time. iv. 109.

Hellenization ' is predieated (for the first time 145a The first traces of this lie very far back.

I think in Greek literature; in the definitely As early as Pherecydes (if it be the fifth cen-

linguistic sense: ko.\ IWrivtaQvaav t\)v vvv tury author of that name) Peueetius and Oeno-

yXoxrarav to't€ irpSnov virb rwv ' ' Afj.it paKiunHiv trus already connt as children ol Arcadian

^vvoiKticravTcev, ol 5« &A\oi 'Afj.<piAoxoi t3dpf3apot Lycaon.

A HISTORY OF THE PELASGIAN THEORY. 209

^ 15. — Ephori j.

If there is one writer who represents for us the characteristics, good or
had, which distinguish fourth century historians from fifth, it is Ephorus of
Oumae. The pupil of Isocrates, he was brought up in the laxest sect of the
pijropes ; and the fragments which we have of his work show how in-
dustriously he improved on the historical method of his master. Not only
was his work on the early age of Greece the first and the most copious of
the fourth century redactions, but it has been shown by more than one
modern writer practically to have held the field until far on into the
Alexandrine Age ; to have been a standard book of reference for Polybius,
and to have supplied Diodorus with almost the whole framework of his
history for this period. Strabo, too, quotes him repeatedly on points of early
ethnology.

It is from Strabo that we learn, among other points, that Ephorus had a
Pelasgian theory of his own. In the well-known passage 146 in which Strabo
summarizes the views which had been held by Greek writers on this matter,
a large proportion of the more important data are assigned to Ephorus by
name ; and the whole of the Homeric evidence is marshalled in a form which
makes it highly probable that we have here an abridgement of Ephoran
commentary : for phrases characteristic of the Ephoran theory recur, as we
shall see, throughout it, This theory of Ephorus may be summed up in a
sentence. The Pelasgians originated in Arcadia and nowhere else ; and sp
from thence, all over Greece and beyond, as military conquerors and colonists, at
a -period which can be dated approximately.

Strabo says that Ephorus got this idea from Hesiod ; and quotes the
actual passage. 147 Now we have seen already that this is the only evidence
preserved to us, down to the end of the sixth century, which expressly con-
nects Pelasgians with Arcadia; that it gives an eponymous Pelasgus ; that
it not merely introduces a factor which is out of accord with the Homeric
data, but had already set people thinking how to explain and justify a
Pelasgian Arcadia; and that it had thus been the source of the temptation
to transfer the phrase YleXaaytKov "Apyos from the Thessalian to the
Peloponnesian Argos, with the disastrous results which we have seen.

The 'Arcadian theory' of Ephorus is introduced, in fact, in contrast to what
Strabo regards as the popular theory {opuoXoyovat d-navTes a-^ehov ti) which
made the Pelasgians dpyalov rt cpvXov Kara tijv 'EXXdBa tt a a a v
eirnroXdcrav teal p,dXiara tt a p a rot? AioXeucri to?? Kara
(-) e a a aX la v. This theory, as we have seen, was current from the end
of the sixth century to the days of Herodotus, and was based partly on an
imaginative interpretation of the language of the Catalogue, partly on the
discovery of the place-name UeXaayiwTis. But it sank into very minor

140 Strabo, 221. I.v saon : see p. l v, '> above.

117 Fr. 68, in which Pelasgus is the father of
U.S. — VOL. XXVII. P

210 J. L. MYRES

importance in Herodotus and Thucydides, who both tend to regard Thessaly
as the starting point rather of Hellenes than of Pelasgians. We must infer
however from Strabo's words, that after the eclipse of Herodotean history this
' Thessalian theory ' revived ; and this is indeed abundantly clear from the
writers of the period between Ephorus and Strabo himself. We may fairly
infer, meanwhile, that Ephorus did not hold this theory, or regard the Pelasgians
as the generic aborigines of Greece ; and that in particular he opposed the
' current view ' that either the UeXaayi/cbv "Apyo?, or the YleXao-yLWTis, or the
Xdpiaa of Thessaly were among their primary abodes.

Next, Strabo's argument treats the Homeric passages similarly, but more
explicitly: Kal yap tP/s Kpr]TT)<; enrol kol yeyovacrw, w? cpacnv "Ofujpos,
quoting Odyssey 19. 177 ff. But Homer does not say that the Pelasgians of
Crete eiroiKOL yeyovacriv, and though, as we have seen, it is very probable that
they did ' come to reside in addition to ' its other inhabitants, — ; fas est et ah hoste
doeeri — the Odyssey gives no direct support to this view. The phrase eirotKot
yeyovaaiv in fact, shows that what Strabo is giving us is somebody's
explanation of how Pelasgians came to be in Crete at all : namely that they
were intruders here, just as they were everywhere else but in Arcadia. Who
was this somebody ?

Further evidence follows, about the UeXaayiKov "Apyo? of Iliad 2. 081 :
Kal to Tlekacryucov "Apyo? i) (derraXia Xiyerai. This also is not true, at all
events in the text of Homer which has come down to us. First, Homer
never mentions Thessaly by name at all. Next, as we have seen already,
the Homeric phrase to UeXaayiKov "Apyo?, refers only to that part of
' Thessaly ' which includes Halus, Alope, Trachis, Phthia, and ' Hellas ' in the
narrowest sense : it is the country of the Myrmidons, and the kingdom of
Achilles ; and it does not include even places like Phylace and Pyrasus, 148 much
less the head of the Pagasaean gulf, or the country round Tricca or Larisa.
This Thessaly, in fact, winch, as Strabo goes on, includes to fieTagv twv
ifJi[3o\o)v tov Yirjveiov Kal TOiv %eppuO"KvXwv e<w? rfjq opeivf)? Trj<; KaTa Yllvhov
is the Thessaly, not of Homer, but rather of Aeschylus ; and the reason
why it is either ' Pelasgian ' or ' Argos ' is the same also as in Aeschylus ; — 8ia
to i 7r a p % a l TOiv TOTiwv tovtwv tovs UeXaayov^. It is an iirap^ia, an
' annexation' of the Pelasgians, not their original home.

Here, again, as in the previous instance, what Strabo is reporting is
somebody's views about Homer, and about Aeschylus also ; and this somebody
has catch-words of his own, eiroucoi, iirdp^ai, arising from his theory and be-
traying it whenever they recur.

A few lines below, Strabo refers again to Ephorus by name, ascribing to
him the use of YIe\ao~y£a as a name for Peloponnese. From this, we may be
pretty sure that Ephorus also, like that early fifth century school of logo-
graphers which Herodotus and Thucydides ignore, took the phrase 11 eXaayiKou
"A.0709 as referring primarily to the Peloponnesian Argos, adjacent to
' Pelasgian ' Arcadia ; and as referring only secondarily to the Thessalian

143 II. 2. 695.

A HISTORY OF THE PELASGIAN THEORY. 211

district. If so, Thessaly was for Ephorus, as for our anonymous 'some-
body,' merely an 'eparchy ' of the Pelasgians of Arcadia.

This impression is confirmed by the words which Strabo adds next, tcai
EvpnriSi)? iv 'Ap^eXdco cf»)a\v on Aavaos 6 TrevTijKoura duyarepwv -rrcmjp
came to Argos and gave the Danaan name to the Ue\acryia)Ta<i oivop,ao-p.evov<;
to 7rptV. 148a This is good fifth century belief, for we have it almost verbatim
in Herodotus. 141 ' It refers of course to the Peloponnesian Argos, but it is
noteworthy that both Herodotus and Euripides make use of the peculiar ethnic
Ue\aayio)n]<;, -wtk>, which only occurs otherwise, in fifth century literature,
as the name of a Thessalian rerpap^a ; and this passage is in an excerpt
from Hellanicus. But why bring in Euripides and Aeschylus in the middle of
this discussion of Homer? Clearly because, not Strabo, but the anonymous
'somebody,' whose views are being traced in contrast with Homer, as with
the airavTe? and iroWoi above, was concerned to claim their support.
And if so, this somebody must have been at work not earlier than the date of
the Archelaus of Euripides. This limits the range of our enquiry a good deal.

Similarly, Strabo goes on, in regard to Dodona : top 8e At'a rov AwSovaiov
avrbs 6 7roi?7T?;? 6 vo fidget Ylekaayixov (quoting Iliad 1G. 233) . . . iroXXol he
/cat to, WireipoiTiKa effvij HeXaayiKa elp^Kaaw. Here again the phrase
auras 6 7roi7]T7j<i ovopui^i has all the look of an attempt on the part of
'somebody' to claim the reluctant Homer and the others who called the Wild
West ' Pelasgic ' as supporters of his theory that the Pelasgian hegemony, more
or less forcible in its extension, had reached as far as Dodona, if not even into
Epirus — &><? Kal fie%pi hevpo eirap^dvTw v — and out comes the catchword
again. Now this exactly accords with the known views of Ephorus about
Dodona: for Strabo says of Dodona in another context 150 can h\ co? ^aiv
v E0opo9, TleXaayoiv Xh p v [xa. Zeus of Dodona, that is, is HeXaaytKos per se
and Aa)8(i)valo<; per acciclens : as fine a rhetorical inversion of the Homeric
phrase as could well be devised.

By this time, I think it will be clear that the anonymous fourth century
' somebody,' whose views we have been tracing in this passage, is none other
than Ephorus himself; and that what Strabo is giving us is a detailed analysis
of the Pelasgian theory of that writer, quoting him by name only wheu his
views diverge from those which were orthodox in Strabo'stime — which is very
seldom — and quoting authors earlier than Ephorus only when their testimony
is either of crucial value, or had required special ingenuity to make it
' fit in ' with the theory.

We begin also, I think, to see the connexion between the curious ami
detailed commentary on the Homeric evidence, on the one hand, and the
Statement which follows immediately, that the ancient author who really best
supported the Ephoran theory, ami indeed suggested it to Ephorus, was
Hesiod. It was indeed a choice between irreconcilables. The learned world
from Acusilaus to Thucydides had expended itself in constructing theories

l48a Fi. 227 : already noted above, p. 191. I50 Strabo, 327. The dependence of this mi

u '' ii. 171. the Hesiodic neKacryHy ibpavov is obvious.

p 2

212 J. L. MYRES

about the Pelasgians which would fit the Homeric evidence as they under-
stood it ; but one group of early passages had stood out, and could not be
made to fit. These were the statement of Hesiod that a personal and there-
fore primeval Pelasgus was the father of Lycaon of Arcadia, and the
conformable witness of Asius that Pelasgus was Earthborn and the First Man.
Pherecydes, on the other hand, had collected round the passage of Hesiod a
mass of local genealogies which went back to Lycaon ; and he had probably
been led to connect with these Arcadian genealogies the barbarous North-
west round Dodona, and places as far beyond as Peucetia and Oenotria. Mean-
while Acusilaus and Hellanicus had tried to reconcile the Homeric and the
Hesiodic schools, by applying to the Peloponnesian Argos, with its citadel
Larisa, the Homeric phrase about to HeXaayiKov "Apyos in Thessaly, and also
the Homeric statement that some Pelasgians (who however had nothing to
do with to HeXaayi/cov "A/r/o?) dwelt round a place called Larisa. The
tragedians belong wholly to this popular syncretistic school. Herodotus and
Thucydides, on the other hand, use mainly Homeric data, but supplement
these by fresh search for objective fact, and by new methods of interpretation.
But now the reaction from anthropology, which Thucydides had foreseen,
has come ; and it is entirely in accord with the methods of fourth century
rhetoric, and with the known bent of his own genius, that Ephorus should
nppear in due season with the mission to construct irpo^ to irapa-^prjpa
dicoveiv a completely inverted pyramid, resting its slender apex on the one
outstanding passage about a personal Pelasgus in Arcadia, and incorporating
the Homeric passages, somewdiat unsuccessfully, very near the broad end
of the structure.

With this clue in mind, the rest of the passage of Strabo is instructive
reading. The remaining passage of Homer, about the ' actual ' Pelasgians
among the allies of Priam, 151 is dismissed in a fashion as brief as it is charac-
teristic: kcu Tot9 iv Tfj TpeoaSi K.iX.L^iv'Op.^po'i elpi]/ce Tou? o/xopovq UeXaayoix;.
Now this, once more, is simply not true, unless the Homeric text has suffered
grievously since Strabo's time. Moreover, if it were, it would make Homer
group with the Pelasgians just those allies of Priam who are least 'at home '
in their Homeric position on the map, when compared with the historical
Oilicians ; and so would afford the plainest suggestion of to TroXvirXavriTov}^

That the Aeschylean theory, too (however well it suited Ephorus in
Thessaly), needed amendment in Pelopounese, is clear from the adversative
clause which follows. AlaxvXo<; he i/c tov rrepl ^lv/c7]vas"Apyov<; cprjalv iv
'IfceTioi /cal Aavaiai to yivo<; avTOiv xa\ ttjv YleXoirovvria~ov Se UeXacrytav
4>r}alv"E<f)opo<i KXijOfjvai; and then follows the quotation, already noteil, from
Euripides. Aeschylus, that is, was in error in supposing that it was because
the Peloponnesian Argos was HeXaayt/cov that Peloponnese was called
YleXaayia ; and Ephorus has set him right. For it is not merely the XleXaayi-
kov "Apyo? of Argolis, but Peloponnese as a whole, which on his theory

II. '1. 843. that of Herodotus. In the fifth century it is

1,1:1 We may note in passing the marked anti- the Dorian Hellenes who are the migratory
between the ethnology of Ej>horus ami i-ironcm of Greece, Tr<>\vnAavr)Tot Kapra. (i. 56).

A HISTORY OF THE PELASGIAN THEORY. 213

acquired the name UeXaayia ; and it acquire! it, as we have seen, not in
pre-Danaandays from the Argive YleXao-yo? of Aeschylus— wide reachii
h\< eirapxiat were — but from the Pelasgian arpancoriKOL of Arcadia.

§ 15. — Tht Successors of Ephooms.

Two classes of data, it will be observed, have evaded, hitherto, the wide-
spread net of the new 'Arcadian Theory' : they will have to form the very
cornice of the inverted pyramid ; and they are just the data which had most
contributed in the fifth century to throw fresh light on the realism of the
Homeric evidence. We have not, in fact, had a word, as yet, either about
Lemnos and Imbros, or about Attica.

Strain, goes on however (with an adversative consturction once more)
'AvTirc\ei8>i<; Be irpcoTOV? cprjaiv avrovs to. 7repl Aijfivov ku\ "lp,/3pov fCTicrai,
xal Si) tovtcov nra? Kai fiCTU Tvpptjvou Tou "Atuo? ei? Ttjv 'IraXlav a-vvapai.
Now it is not very likely that any of Anticleides' writings were extant early
enough to be of use to Ephorus ; and Philochorus, the Atthidographer who is
particularly responsible for the speculation about the UeXaayol-UeXapyoi
which Strabo quotes next, is even later still. It follows that what Strabo
is doing now, is to supplement and develope the theory of Ephorus from the
works of his immediate successors. In both cases the Pelasgians are repre-
sented not as aborigines but as immigrants ; but the verbs are no longer
eiroiKeh' and iirap^ai, but fCTicrai and kirKpoirav. 'How exactly the fact of
these Pelasgian settlements was worked into the general structure of the
theory, there is nothing in this passage to show ; but the silence of Strabo as
to Ephorus, and his use of later writers to supplement his theory on these
two points, certainly suggest that a difficulty had been felt. In the case of
Athens the problem was simplified in advance for Ephorus by the circumstance
that, as Herodotus observed, the Athenians, whatever their origin, were so
thoroughly Hellenized as to be reckoned Toiat, irpwroiat \eyop.evoiai eivai
'EXXrjvcov <ro<f)Lr}v, 152 and therefore furthest removed from the simplicity and
folly of barbarians. If, that is, the theory of Ephorus arose as a false corollaiy
from a cultural classification of extant Greeks, such as was contemplated in
the time of Thucydides, the Athenians must at once have fallen out of the
list of possible candidates for genuine Pelasgian ancestry ; and if so, the stories
in Hecatanis and Herodotus about their dealings with Pelasgian irXain^ai
would come in as proof of the early date of Attica's conversion to Hellenism.
The philological speculations of Philochorus about UeXaayol-lleXapyoi rest on
inadequate knowledge of the history of the Attic dialect. But, whatever their
validity, they are incompatible with any theory which did not reject or more
probably ignore) the whole of the Herodotean treatment of the 'aboriginal '
Pelasgians of Attica, and lay stress solely on the Herodotean admission that
certain Pelasgians 'came and went' between Attica and Lemnos.

'"- lhlt. i. 60.

214 J. L. MYRES

In the case of Lemnos we have further evidence of the fourth century
treatment of Homer. Homeric proof of the late arrival of the Pelasgians in
Lemnos existed indeed, though only of a negative kind, and so far Anticleides
was justified in asserting that the Pelasgians were not aborigines but colonists.
But in laying stress on the negative evidence, he ignored the positive
testimony of the Iliad to a jore-Pelasgian ktI<ti<; in Lemnos; and it was only
by so ignoring it that he was able to state ir p cot o v 9 . . . aurou? rd irepl
\i)p.vov /cal Ififtpop KTLcrai.

§ 16. — Pelasgians and Tyrrhenians.

Strabo's citation from Anticleides introduces another new feature, when
it attributes to Pelasgians of Lemnos and Imbros a share in the foundation of
Etruria. Attempts at an explanation of the western Tyrrhenians by means
of a Pelasgian theory of the Aegean go back, as we shall see, at least as far as
Hellauicus ; and both Herodotus and Thucydides mention 'actual' examples
of the two peoples in an association so close as to border on identity. But
the statement of Anticleides is, I believe, the earliest which connects the
' actual' Tyrrhenians of Etruria with any part of the Aegean where ' actual '
Pelasgians existed in historic times. It is on this ground that I have
reserved till now an examination of the literary evidence about the
Tyrrhenians by the same method of criticism to which I have confined myself
in the preceding sections. If it leads to an intelligible result in this case
also, I think I may claim this as some confirmation both of my previous
results and of the method itself.

Considering how much has been written about the Tyrrhenians and how
large a place they filled on the Greek horizon, it is almost surprising to find
how little early evidence about them has survived in Greek literature.
Homer has no mention of Tyrrhenians at all ; and the isolated passage in
Hesiod's Thcogonia (1. 1016) is suspect. In fact the only direct reference in
literature earlier than the fifth century, is that in the Homeric Hymn to
Dionysus (1. 8). Here the sea-pirates who kidnap Dionysus, and are
miraculously punished by him, are introduced without comment us Tvpcnjvoi.
But the Hymn gives no internal indication of the date or place of the
episode, except that in 1. 28 Egypt and Cyprus on the one hand, and the
Hyperboreans on the other, seem to lie on the poet's horizon ; and this does
little but confirm the conclusion suggested by style and language that the
Hymn may belong to the sixth or seventh century, and not much earlier.
At two points in the Hymn there may be traces of ' Tyrrhenian ' proper
names ; but if there are, they are hopelessly corrupted. It is possible, but is
not proved by anything in our text, that the Hymn may belong to the same
Cycladic cult of Dionysus as the fragmentary Hymn I, with its allusions to
Naxos and Icaria; I55S but it is also possible, as the unexplained allusion

153 The earliest version of tho story of Dionysus, which implies this is that in Apollod. iii. 5. 3.

A HISTORY OF THE PELASGIAN THEORY. 215

to the bear 1 "' 4 suggests, that it may belong to the Brauronian cult: in
which case this Hymn (or the legend which it embo lies) may be the source
from which the Tyrrhenian nam.' came later into the story of the Pelasgian

raids round Attica.

Jn the fifth century four distinct stories were told about Tyrrhenians m

the Aegean basin. 155

(1) Herodotus 1 '" 1 and Thucydides 157 are agreed that Tyrrhenians existed
still, in the fifth century, in the district enclosed between Chalcidice, the
Strymon, the Axius, and the inland Paeonia ; and that .they were adjacent
to (Hdt.j, if not actually part of (Time), the Pelasgians who survived in that
district. Thucydides adds, as we have seen, that they retained a language of
their own, and connects them with certain inhabitants of Lemnos and Attica
who seem to be those whom Herodotus calls Pelasgians. But neither writer
connects these actual fifth century Tyrrhenians with the Tyrrhenians of the
West. i:,s

(2) Sophocles is quoted 159 as having used the double phrase kul
'tvpavvolai UeXaayok of a part (or the whole) of the people of the pre-
historic realm of Inachus, namely the Peloponnesiau Argos. But we have
seen in the case of the word YleXaayoi, first, that its application to the
Peloponnesiau Argos results from misinterpretation of the TleXaayucbv
v Ap 7 o9 of Homer; secondly, that already in the time of Aeschylus this
prehistoric realm was regarded as including a large part of central and
northern Greece, and particularly the Thessalian Pelasgiotis. There is
nothing in the Sophoclean use of ' Tyrrhenian ' to preclude this interpretation
of the passage, and there is no suggestion anywhere that there either were or
had been ' Tyrrhenians ' in the realm of Inachus in any other sense than that
in which there were or had been ' Pelasgians.' The passage in fact is only ot
interest as confirming the evidence of Herodotus and Thucydides as to a
growing belief in the fifth century that the ' Pelasgian'' and the ' Tyrrhenian
names" went together ' in some way; and, as we shall see shortly, by the
close of the fourth century these names had become practically inter-
changeable.

(3) Hellanicus, 160 though he does not expressly mention Tyrrhenians
in the Aegean, has a theory about the origin of the Tyrrhenians in the West
which derives them from his Pelasgians of Thessaly. These Thessalian
Pelasoians, on being expelled from Thessaly by the Hellenes who. for

i See 1. 16 and Crusius' note. B rodotus, and therewith that of Thucydides

155 In( ■ tgedians' use of T W «^ that E. Meyer is enabled to con

5t0 ck Epithet of £x«* or ««.r, Aeieh. He, ennl Tymner un B

K 567, Soph- -4*. 17. tegaischenM « i imcht/ Forschungen i. p. SI.

159 Fr. 256=Dionys Hal. i. 2d.

i. ui.

,- j' y jog M« Fr. l=Dionys. Hal. i. 28.

i; > 3 It is only by rewriting the passag< of

216 J. L. MYRES

Hellanicus, seem to have been immigrants from somewhere) 1C1 took ship and
landed in Italy eVt S-rrivrjTi Trorafxtp, that is, on the Umbrian coast near
Spina ; 162 they then went up country eU Kporcova tt6X.lv.

That Hellanicus however had himself no evidence of the existence of
the Tyrrhenian name in Thessaly, is suggested by his use here of the Pelasgian
name solely, so long as he is describing events in Thessaly or indeed anywhere
outside Italy ; and by his statement that it was only on arrival in Italy that
the Pelasonan refugees took the name ' Tyrrhenian.' i0:J At the same time we
must note that elsewhere 104 he ascribes a settlement at Metaon in Lesbos to
one MeTa? Tvpprjvos ; and as most of the Lesbian towns were of Thessalian
origin there is a prima facie case for regarding this Tvppijvos as coming from
thence. 165 He might however have been a Pelasgian from Lemnos or the
Hellespont.

Hellanicus gives elsewhere, as a lower limit of date for this migration,
the third generation before the Trojan war, and the twenty-sixth year of
Alcyone, priestess of the Argive Heraeum ; and Philistus, a little later, 106
gives the same date, in the formula ' eighty years before the Trojan war.' In
both cases the actual date in question is that of the expulsion of the Sicels
from Italy into Sicily ; but as the Ligurians, who according to Philistus
expelled them, were themselves under compulsion from ' Umbrians and
Pelasgians,' the presence of Pelasgians in or near Umbria is presumed at a
date not later than the Sicel migration. We cannot however be certain that
the Pelasgians who landed at Spina were the only people of the name whom
Philistus (or even Hellanicus) believed to be at that time in Italy.

(4; About the Western Tyrrhenians however Herodotus has a quite
different story, which he gives on Lydian authority 1G7 : namely that they are
of the same origin as the Lydians. His story is that in a time of famine
these Lydian Tyrrhenians took ship and ' after passing many peoples' came
to the 'OfiftpLKOL where they founded cities ; and there they live still. They

161 The story added by Dionysius, that this lanicus, but not explicitly so), does not seem to

happened in the days of Deucalion, cannot be prove Tyrrhenian, or Pelasgian, or even Thes-

traced to any early source. The nearest analogy salian origin. That the latter view at all events

is Herodotus' statement (i. 50) that in the day> was popularly believed later is confirmed by

icalion the Hellenes oIksov yrjv t^v *0»«tw the analogy of Ravenna. But even a Thes-

and in the time of his grandson Doras migrated salian origin does not prove that the colonists

to the Histiaeotis below Ossa and Olympus; were either Tyrrhenians or Pelasgians, and

but this does not prove that in the intervening Btrabo (214), who is our authority for this, has

generation they occupied the intervening terri- chosen to describe Spina as it6.\ai 5e 'e\\t\v\s

tory, though Dionysius very likely thought it iroAts evtiotos, which is bad for its Pelasgian

did. Hesiod {Cat. fr. 11) and Pindar (01. 9. origin.

64) seem to have regarded Deucalion as king of 1,;! There is some late evidence for a belief

Opuntian Locris, or at least of Opus ; but we do that then were Pelasgian settlers in Lesbos:

not know how early it was discovered 'that this see especially Strabo, 221, 621, Diod. 5. Si,

king of Opus was the invader of Phthia. Plin. X.If. 5.31. o9.

; - That Spina should have maintained lliJ Fr. 121 = Steph. Byz. s.v. Merawv.

tributary relations with Delphi down to the 165 Dionys. Hal. i. 22.

of its destruction, as is stated by Dionysius J,, ' ; Dionys. Hal. I.e. — fr. 2.

(i. 17, perhaps also on the authority of Hel- ,li: i. 94.

A HISTORY OF THE PELASGIAN THEORY. --'17

got their name from their leader Tyrsenus, who was son of Atys and conse-
quently (i. 7) brother of Lulus the efionymos of the Lydians. Here the chai
of name suo-o-ests the same conclusion as in the case of Hellanicus, namely
that Herodotus had qo evidence before him of a Tyrrhenian people in Lydia.

( h, the other hand a dichotomy of the Lydians, such as his story presumes
is in accordance with a native Lydian tradition of fifth century date : for
Xanthus the Lydian 168 gives, as the sons of Atys, Lydus and Torebus (or
Torrhebus) and adds that the languages of their respective descendants stood
to one another as Ionic to Doric, that is, they were closely-kindred dialects.
Xanthos however gives no indication of a Torrhebiau emigration; hut he
knows of a town Torrhaebus in Lydia. Not own 'Tyrrhenus' bowever is
known either to Herodotus or to anyone else. 1 '''.

Another point is perhaps worth noting', to complete the parallel between
the accounts of Herodotus and Xanthus, and to suggest a line of argument
which may very likely have been present to the mind of the former. Hero-
dotus introduces his account of the Tyrrhenian emigration as a footnote to
the Lydian invention of vraiyvtai, which he ascribes to the Lydians, on Lydian
authority, in a passage the rest of which is remarkable for its detailed
knowledge of things Lydian. 17 " Now we do not know enough either about
Lydian or Tyrrhenian, or even about Hellenic Traiyviat, to be able to confirm
or to dispute Herodotus' account ; but we may fairly assume that in his time
there was actually sufficient similarity between these pastimes, to uphold such
a story; and further that such similarity between Lydian and Tyrrhenian
games was one of the testimonia to the story of the Tyrrhenian emigration—
as one might argue from the games of New England or Virginia nowadays.
So that it becomes important to note that in Xanthus also 171 the place
Torrhebus has a local culture-hero named Carius, who is inventor artium, and
that is why Lydian music, in particular, is called Torrhebian : for here we
seem to have another phase of the same general story of a Lydian or Tor-
rhebian culture-hero.

It is by this time fairly clear how Herodotus came by his story, at
its Lydian end. At its Italian end the story is clearly a variant of that of
Hellanicus: for ' TTmbria ' in Herodotus 17 - extends northwards as far as the
foot <>f the Alps, and so includes the site of Spina. Meanwhile his phrase,
edvea iroWa Trapafxet^afMevovs, looks very like an attempt to summarize a
long series of data as to 'Tyrrhenian 5 settlements, or attempted settlements,
on the route between Lydia and the head of the Adriatic.

Summing up the evidence of the fifth century writers we reach the
following presentation of the fifth century view of the Tyrrhenians; and we

168 [.',. i DioilVS. Hal. i. 'JS. l; " i- 94 <paal 5e avroi AuSol ical Tas naiyias

"• : * The name Tvppr)v6s would I"' a natural Tas vvv ff(pi<Ti t« koi "EWllfft KaTeaTeurras

'ethnic' if there was ever a place called Tyrrha, £«ot«v i&vprifia yeviodai- aua 5e Tanas re

and it was believed in quite late times that ^ei/peflJ^ai ira^a a<picn \iyovat xal Tvpayvl-nv

ili,!, was mi, h a town in the South Lydian ai-our/irar »5e irtpl avrSu xiynvres.
district of Tonhcbia (Et. Mcuj. s.v. Tipawos) 1:1 Fr. 2, summarized by N

and that Gyges came from thence. Bui this Damascus,
pvovi for the fifth century or earlier. '•"- i\". >'•'■

218 J. L. MYRES

note at once the remarkable likeness between its main features and those of
the Pelasgian theory at the same phase.

First, there are 'actual' Tyrrhenians (1) north of Chalcidice, (2) in
Etruria ; but no fifth century writer has recorded any attempt to identify
them.

Secondly, the ' actual ' Tyrrhenians of Chalcidice are closely associated
with ' actual ' Pelasgians in our two best authorities.

Thirdly, speculation has been at work, connecting, on the one hand the
'actual' Tyrrhenians of Chalcidice with the intrusive Pelasgians of Lemnos
and Attica, on the other hand the ' actual' Tyrrhenians of the West (1) with
' theoretical ' Tyrrhenians in Lydia, now extinct, (2) with ' theoretical '
Pelasgians in Thessaly, also extinct now.

Fourthly, in popular belief, represented by Sophoclean Tragedy, the
name ' Tyrrhenian', again in the closest association with 'Pelasgians' has got
a general connotative sense of ' pre-Hellenic in the Aegean,' which exactly
corresponds with the behaviour of the Tyrrhenian individuals whose exploits
have come down to us in our one epic source, the Homeric Hymn to
Dionysus. 173

But no sooner do we pass from the fifth century into the fourth than
all is exaggeration and confusion.

First, as we should expect, the connotative use of ' Tyrrhenian ' to mean
' violent and piratical ' crystallizes into a definite theory, assigned to Ephorus
by name, in which the Tyrrhenians play almost exactly the same part at sea,
as has been assigned to the Pelasgians on land. The crucial passages are :

(1) Strabo 410, where rCphorus accounts thus for the lateness of Hellenic
expansion in the West, 174 tovs yap 7rp6repov SeSiivai ra Xr)cni]pia ra>v
Tvppy]vwv Kal tt)v o)p,oTy]Ta rwv ravri) (3apj3dpu>v. Here the Tyrrhenian is
the type of Outland barbarism, as the Pelasgian is of pre-Hellenic barbarism
in the Aegean.

(2) Strabo 477, where the writer, speaking of the Cretans, says f.terd
toi/9 Tvppi]vov<; o't /uudXiara ihnooaav ttjv ku6' i)/xa<; OdXarrav, ovrol elaiv ot
hLahe^afxevoi rd Xyanjpia. At first sight it is tempting to take this as
referring to the Cretan piracy of historic times, which is much in the mind
of Strabo himself. But if the ascription to Ephorus is correct, this is
out of the question, for the Cretan piracy did not appear to be serious
till after the age of Alexander. Another possible interpretation would
be to regard rd Xyarrfpia — a regular Ephoran catchword, like eiroiKOi and
e-rrdp^ai — as the victims' expression for a ' sea-power.' But there is no
evidence that Ephorus was acquainted with the Thalassocracy List which

; This is all quite independent of the late l7 * lie assigns the foundation of the western

and far too sweeping generalization of Dionysius Naxos ami Megara to the fifteenth generation

(i. 25, on tli" passage of Sophocles)* Tupprivlas after tin' Trojan war (1184 B.c. [15x30=]450

fxev yap Br) uvoixa rhv XP' )V0V tKtlvov kvh. rrjv =734 B.C.).
'Ewdba i\v.

A HISTORY OF THE PELASGIAIN THEORY. 219

comes to us through Diodorus ; and even if he was, it cannot be argued that
he described as Tyrrhenian the sea-power" 5 which the List calls Pelasgian :
lor the sea-power which succeeds it is not ' Cretan ' in the List, but ' Thracian.
The only alternative is to regard the 'Cretan' ^ar^pca as the famous
' sea-power of Minus,' and to regard the Tyrrhenian Xrjarfjpia as the Ephoran
equivalent for what Herodotus and Thucydides know as the « Carian
sea-power which Minos overthrew. 1 ™ So, whereas in the A,-,an this
'Tyrrhenian' sea-power was broken by Minos, and permitted He lemc
expansion early, in the West Minos tailed as Herodotus knew,, and Hellenic
expansion tarried till the fifteenth generation after the Trojan W ar.

Secondly, whereas Hellanicus had made his Thessalian Pelasgians change
their aame on their arrival in Italy, and so leave the West a free field
for Tyrrhenians, the fourth century, from Philistus onwards, admits unmodified
Pelasgians in Italy. In Philistus' account, already cited, 1 " of the dis-
possession of Ligurians and Sicels southward, their invaders are not
Umbrians and Tyrrhenians, as we should expect from the fifth century
evidence, but Umbrians and Pelasgians. The later writers carry this con-
fusion further, sometimes identifying Pelasgiau and Tyrrhenian, sometimes
distinguishing them. The 'Thessalian' Ravenna, for example, strengthens
itself against ' Tyrrhenian ' attack, by admitting its ' Umbrian' neighbours. '^
In Southern Campania, beyond the Sarnus R. lie elra Tvpp^voi km
UeXaayoc, fieTa Tavra hk ^avvlrar kul ovrot 8' e^eireaov €K tg>v tottcov. '
Diodorus, in fact, was probably under no misapprehension when he said
that 'the Greeks' apply the name 'Tyrrhenian' to Latins, Umbrians,

and Ausones indifferently. 180

Thirdly, the weakness of the evidence which in Hellanicus' story connects
the Pelasgian immigrants from Thessaly with the Tyrrhenians of Emma,—
and perhaps also a discrepancy between the date of king Nanas of Thessaly
in Hellanicus, and that of king Atys of Lydia in Herodotus and m the few
writers such as Timaeus lSl who followed him in this matter,— seem to
have led later to the conclusion that in the West there were two movements
of colonization, one earlier and ' Pelasgian,' the other later and 'Tyrrhenian
A good example of the duplication which ensues is that legend ot Caere, ■
in which a Thessalian-Pelasgian in the town speaks Greek— x«^€— to
a ' Tyrrhenian ' assailant, and is understood by him. 133 The Pelasgian emi-
gration to the West from Thessalv, moreover, was certain sooner or later

"a Placed by Eusebius (Jerome) between 177 Fr. 2 Dion. Hal. i. 17.

1056 and Ml B.C. ; and by myself about a I78 Strabo, "214.

. utury later [J. U.S. xxvii, pp. 88, V26-7). 17S Strabo, '217.

w This agrees well with the fourth century * Foi insl in. cssee the literature ... B

date for the spread of the Hellenes over the and Reinach, Zes Celt* d I

Pelasgian 'eparchies' of the mainland : for Ion du Danube, 1894, pp. i 1 "•

son of Xuthus is very nearly contemporary Fr. 19.

with Theseus, and Theseus is one generation Strabo, 220.

below Minus and one general tin '«< That Caere, or rather the uurefonned

Trojan war. llellen therefore was fuur genera- Agylia, had like Spina, regular relations «ith

tions before the Trojan war, and Xuthus ; ,nl Delphi, and even a treasury there, pro

Doru> were contemporaries of Minns of < 'i notliing as I

.1-12111.

220 J. L. MYRES

to be confused with the far earlier movements implied in the genealogy
which Pherecydes constructed for the children of Lycaon of Arcadia. One
version of the latter brought Oenotrians from Arcadia to Italy as its first
inhabitants, and a kindred version (which however only comes to us
through Dionysius, and is not assigned to Pherecydes or his followers by-
name) sets this Arcadian movement as far back as the seventeenth gener-
ation before the Trojan War. The evidence however for this double-coloniz-
ation is all later than the fourth century : 183a it naturally proves nothing
for any period earlier than the circumstances which called the theory
itself into existence; and these circumstances are indicated very clearly
in Dionysius' own version of the story of the Pelasgians in Italy, 184 for
part of which he claims the support of Hellauicus. For he represents
the Thessalian Pelasgians of Hellanicus as beino- themselves a detachment
of the militantPelasgians of Arcadia, who were not invented till a century
after Hellanicus' time ; and he puts their arrival back six generations
before the days of Deucalion, whereas Hellanicus had kept them in Thessaly
until the invasion of the Hellenes, at least one generation after Deucalion,
and ..»nly three generations before the Trojan War. The whole story, in fact,
as viewed by Dionysius, is seen through the spectacles of Ephorus ; or rather
perhaps of some follower of Ephorus whose aim was to work into the Ephoran
theory some part of the calculations of Pherecydes. 18 "'

Fourthly, the Tyrrhenian name became more and more widely applied to
the Pelasgian invaders of Lemnos, Attica, and other parts of the Aegean.
The statement of Thucydides, that his Tyrrhenian-Pelasgian folk in Mt.
Athos were akin to the invaders of Lemnos, lay open to misconception in
proportion as the word ' Tyrrhenian 'gained more generic vogue ; and Ave have
already seen that Hellanicus had placed a 'Tyrrhenian' colony in Lesbos, over
against "AvravBpov r>)v UeXacryiSa. There was some excuse, therefore, for
the attempt of Anticleides to reconcile the accounts given by Herodotus, and
by Hellanicus, of the western Tyrrhenians, by causing Pelasgians from Lemnos
and Imbros (who on Thucydidean authority were akin to the Tyrrhenian-
Pelasgians of Mt. Athos) to join Tijri Ji< '.mis, son of Atys,and his men. eOiea ttoWci
7rapa/jL€t\jra/u.€i'ov^, as Herodotus says, on their way to Tyrrhenia-in-the-West.

But it is quite another affair, when Ephorus describes the Lemnians as
Tyrrhenians without, qualification ; I8 ° or when Philochorus 187 retells the story

'" ' I - i- -' t out in great detail by Ridgeway, to Hesiod, if fragment 85 of Pherecydes was

Early Agt oj Greece i. pp. 231 II'. extant and known to him.

Dionys. Hal. i. 22. lS6 Diod. 10. 19.

'" I have already commented (p. 187, n. 11 lb7 Fr. 5. We have already had two e^peri-

nn the doubt which must exist as to the ■! tte of ences of Philochorus as a philologist and it is

i!i\ statement attributed to Pherecydes, and in the very next fragment, fr. 6, that he derives

my impressiou that, though the earliest of t lie the name of the Homeric Series from <Tlveo6ai

three writers of this name was a \oyoypdcpus with refeicnce to this same raid. But in this

ami probably compiled genealogies, the quota- fragment he calls the raiders Pelasgians. The

tions themselves betray the influence of the Homeric StVnes however as we have seen, have

Ephor; and may lie quite late. It is no more to do with the Pelasgians than they

certainl} remarkable that Ephorus did not have Avith the Tyrrhenians.
acknowledge his debl to Pherecydes as well as

A EISTORY OF THE PELASGIAN THEORY. 22]

of the Pelasgian raid on Attica with details derived from Hecataeus and
Herodotus, but with 'Tyrrhenians' substituted for Pelasgians throughout,
and with the philological moral rvpavvos el'p>]Tac diro twv Tvpp^v^v t6)v
/3iai(ov /cal \r)<TTO)v ij~ dp^i)? . . . Tvpptjvol yap oXiyov riva xpovov
(KKi'icravTes iv t.u? 'AOquais . . . toWoI pblv avrcov uttcoXovto . . . aWoi ol
(stccpvyovTes Aijfivov /cal "\p,(3pov wK^crav . . . and then returned when irapOevoi
«pKTev6p.evai rfj 6e<p were at their mercy at Brauron. After this it is not
surprising that Apollonius of Rhodes, 188 followed by Plutarch 189 and
Polyaenus, 190 should have described as 'Tyrrhenians' the persecutors of the
Minyans ; that Aristoxenus 191 should describe Pythagoras as a 'Tyrrhenian '
from Lemnos ; that Diogenes Laertius 1 '- should describe one Mnesarchus as
Tvpprjvov ovtcl ko.\ yeros twv Anp,vov kol "\pj3pov /cal %/cvpov J^LKliardvTOJV
Tuppijvwv ; or finally that the Lenmians who were conquered by Miltiades
should rank, for Cornelius Nepos, as Carians. 193 Only much later (with the
single exception of one passage of Charax) does the revival of Herodotean
authority permit Stephanus (s.v. 'H^aio-r/a?), Suidas, and Zenobius
(s.v. 'Epfxwvetos x i *P L< >) to recur to the fifth century name of 'Pelasgian.'

The mention of Scyros is particularly instructive, because its inhabitants
had been noted by Thucydides m as Dolopes, of a well-known mainland stock
of ordinary North-Greek type. 195 Ephorus however called them 'Pelas-
gians,' 196 as we might almost have guessed, seeing they are e-rrotKoi from
North Greece ; Scynmus couples them with the men of Sciathos as He\aayoi,
but gives them a quite different origin, e'/c ©pa/o?? Siaftdvres, o>9 Ao'709 ; l
Nicolas of Damascus calls them * Pelasgians and Carians ' 19S and Diogenes,
as we have seen, couples them with the men of Lemnos and Imbros, but calls
all three peoples ' Tyrrhenian.'

§ 1/. — Conclusion.

Anyone who has followed this analysis of the Greek authorities as fai as
the close of the fourth century will agree, I think, that there is not much to be
gained by classifving the unauthenticated statements of the writers further
down. Anyone, moreover, who is familiar with those statements, will
recognize at once how large a proportion of them consists in direct elaboration
of the Homeric and Herodotean connotative view, that 'Pelasgian' meant
pie-Hellenic' in much the same sense as 'British' is popularly used in
England for 'pre-Roman,' or ' Druidical' for ' pre-Christian ' ; and how large a
proportion of the remainder are Te/xri y>7 T °d peydXov heinrvov E cpo p o v.
Take the case of the famous Pelasgian settlement in Rome. There is

• Argonautica iv. 17ti0. Pammon, all he has to say of hint is that he

189 V- ''''• 21, Virt: Mid. 8. 19 ° vii. 49. betrayed a Greek anchorage to the I' :-

> |>. 1. IK viii. l. 2. 196 Diodorus, xi. 60.

" : Milliades2. lw i. 98. 197 Scyranus, 615.

1,J5 Hilt. vii. 132, is.'.. Though he lias Steph. Byz. s.v. %-tvpos.

oi osionin vii. 1S3 to mention a Scyrian mimed

222 J. L. MYRES

an obvious but anonymous culture-hero; so he is Evander (ev-av$p-o<; = rir
I onus) and of Arcadian origin. There is the place-name Palatium ; so it is
a de-nasalized form of Pallantion in Arcadia. There is archaic masonry
upon the hill; so it is a UeXao-yttcov ret^os : and behold! an 'important
confirmatory proof of the Ephoran theory of an Italian ' eparchy ' of the
Pelasgians ; incidentally also a good excuse for Roman intervention in the
affairs of ' Pelasgian ' Epirus and ' Pelasgian ' Greece.

Nor is the case of the Pelasgians exceptional. I have dealt already
incidentally with the Carian Theory which grew up on parallel lines in the
South Aegean, and more fully with the story of the Tyrrhenian name in the
Aegean and in the West. The story of the Leleges is shorter and more
fragmentary; but in its main outlines it hardly differs. In all, there is an
early period, beginning with a time when there seems to have been a real
but evanescent tribe, of limited geographical range, and some peculiarities of
culture ; and ending, between the sixth and the fifth centuries, with a vague
cycle of memories, and a connotative usage of the name. To this, in each case,
succeeds a fifth-century phase in which, while ingenious theory flourishes,
real search for ' survivals ' of backward folk is perceptible. Then comes
the fourth century, regardless of research, reckless of accuracy or scholarship,
infatuated with headstrong theory, to which the evidence (such as it is)
must conform or be ignored; and then Alexandria, stupidly farraginous, but
rehabilitated lately, as we saw to begin with, as 'evidence to the same effect,
perfectly unexceptionable and as strictly historical as the case will admit of.'

J. L. Myres.

APPENDIX ON APOLLONIUS, ARGON AUTICA I. 1021-4.

I have reserved for discussion in an appendix the one passage in which an ancient
author purports to describe an attempt on the part of ' actual ' Pelasgians to gain a footing
on the Asiatic shore of the Hellespont. The passage itself is of late date ; and my only-
reasons for not treating it among contemporary passages are that the personages to which it
refers can be traced back beyond the fifth century ; that the ethnic situation which it
presupposes has already been shown to be presupposed in the Homeric Age ; and that the
incident itself occurs in a context which links it at latest with the Ionian colonization of
Propontis, and at earliest with the Argonaut-saga, which we know from Homer to have been
current in some form or other before the composition of the Odyssey.

The anecdote in question is as follows. The Argonauts, after passing the Troad,
landed on the Asiatic coast of Propontis, made friends with the Doliones and their
king Cyzicus, and fought some yrjytvUs from the interior, who tried to blockade the Argo
in the so-called x vT ' )S ^'A tl 7 1 ' a ' Cyzicu9. Soon after, they were forced by stress of weather to
put back to the same friendly coast. Then follow the crucial lines : —

l. I Oil -4. <iiht tis aiTi/f vi,<ruv errKppabtws tvorjuev

eppcvai' ovd V7T<> vvkti Aokioves tty aviuvras
ijpuut vrjpfpres t'nT]iaav' oXXd ttov avdpcoi/
MctKpifwv furciPTO IleXacryiKov ape a KeXtrai.

A HISTORY OF THE PELASGIAN THEORY. 223

So there was a fight at <toss purposes, ami great slaughter of the Doliones, and in that
fight \va> Cyzicus slain, their king : whose tomb remained at the city of Cyzicua in

Hellenic times, honoured .-till with Argonautic <"e6X(i.

Now granted that all that Apollonius knew was the foundation-legend of Cyzicus, and
some previous version, not necessarily early, of the Voyage of the Argo : granted also that
the foundation-legend itself was mainly aetiological, and that every self-respecting town in
Propontis, and beyond, had its own ' reminiscence ' of the Argonauts, to prove its antiquitv ;
yet nothing of all this explains either the specific name of the Mu/cpieWj or why the phrase
TltXaa-yiKuv apfa is applied to their raid.

This name, and phrase, completely puzzled the very learned scholiast of Apollonius.
He seems to have begun by applying it to the Doliones themselves; and he explains
(1) that the Doliones are colonists from Euboea ; (2) that Euboea was once called Maicpis
'I.on- Island'; (3; that as Euboea lies ' near Peloponnese, which is Pelasgian,' Cyzicene
(i.e. paullo-post-Euboic) warfare was 'Pelasgian' likewise. 198 We have clearly to look
further than this for an explanation.

Apollonius himself shares, as we have seen,-' 00 the misapprehensions of las time as
to the relations of Pelasgians and Tyrrhenians; and he is therefore not the most likely
person to have held consciously a Pelasgian theory, or recounted willingly a Pelasgian
anecdote, which presumed a quite different view from anything which had been held
since the fifth century, if even consciously so late as this. It is therefore the most
notable, if he has preserved such an anecdote ; and if he has, there is a fair presumption
that he did not invent it, but found it in existence and used it.

The version of the same incident which is given by Apollodorus - 01 su^o-ests that
there was more in the authority which Apollonius was following, than he chose to in-
corporate in his Argonautica. The passage is worth quoting in full: «pj Arjpvov 8e
TVpoo-l(TX ov<JI - -ioKiocri, hv tjaalXtve Ktfi/coy ovtos airoi's vTTt8e^aro cptXorppovus. vvktos
dva\6tvTes (vrtvOev, KM TTfpcTTfat'wTfS uvtittvoUus, dyvoovvTfs nd\u> tois AdXioji Trpoaicrvova-iu.
oi S« vopl^ovre s Ile\a<ry i kov fii'Oi arparov (eri^oi/ yap v tv 6 II e A a crv w i>
crvv e \a) s tt o A e pov pe v o i) pd^v ttjs vvktos (rvvdnrovaiv, dyuoovvres npos dyvoovvras.

Who were these Pelasgians by whom the Doliones were ' incessantky raided ' \ Thev
can hardly be the Herodotean Pelasgians of Placie and Scylace ; partly because the
Pelasgians are apparently still an European people in Homer, and had certainly not yet
reached Lemnos in the Homeric Age ; but still more because it was a sea-borne raid which
convinced the Doliones that the invaders were Pelasgian, and the Pelasgians of Placie and
Scylace were on the same side of Propontis as Cyzicus itself.

But were they Pelasgians from Lemnos? Certainly not, in a poem by Apollonius, or
we should surely have heard something of this exploit in his version of the Lemnian episode.
Moreover, even if Apollonius had thought that there were Pelasgians in Lemnos in
Jason's time, there is Homeric authority, as we have seen, to the contrary.

The whole question is somewhat complicated by the fact that there was also great doubt
in antiquity as to who were the Doliones. Stephanos says'-"- that Homer applied this name
to the inhabitants of Cyzicus wy tovs top "lapapov Kikovcis: but this does not occur in our
text of Homer. It suggests however that there existed some ' Homeric ' source of tradition
about Cyzicus ; and this we shall soon see to be probable otherwise.

In the ordinary way Cyzicus counted as a colony of Miletus; but we know from
Hecataeus of Miletus,- ' 5 at the close of the sixth century, that it had a previous existence
as a town of the Doliones (or Dolieis as Hecataeus himself wrote the name). The
geographical situation is discussed fully and clearly by St raho.-' 1 But who the Doliones
were is only known from one phrase of Kphorus, and from Alexandrine or later writers ;
and opinions differed then in an instructive way. Ephorus 203 describes them as TltXatryovs

199 Schol. Ap. Eh. i. 962, 1024. -"- Steph. Byz. s.v. Kv{ikos.

200 Above p. 221, n. 188. L " : < Fr. 204=Steph. Byz. s.v. AoA.'oves,
- ul i. 9. 18. 1. Apollodorus wrote circa Xi Str. ;"75.

140 B.C. Schol. Ap. Bh. i. 1037.

224

J. L. MYRES

ovras Kai e^Bpcoftws 8ia.K€ifievovs irpbs tovs tijv QerraXiav xai Muyrpjo'iai' kcitoikovvtcis Sia to
arT(\acrdr]vai vV avruv. i.e. he regards the Doliones as exiled Thessalian-Pelasgians ; and in
this he is followed by Conon, 206 who adds that Cyzicus was their king in Thessaly, and
that those who expelled them were Aeolians. It is therefore not merely a confusion of
names when another late writer- 07 calls the Doliones 'Dolopes'; for we have seen
already 203 that the population of Scyros which for Thucydides was ' Dolopian ' had become
' Pelasgian' already for Ephorns, and remained so for Scymnus and Nicolas.

But there was a quite different account of the Doliones, 209 which described them not
as exiles from Thessaly, but as cittoikoi 6erraX&>i>, and consequently kinsmen of Jason, and
fellow-enemies both of the expelled Pelasgians, and of the yr/yevea of the Cyzicene interior.
This was the view of Deiloehus, whom the Scholiast says that Apollonius was following. 210
These yrfyevies are an additional element of complexity in the story. Apollonius says that
it was they who tried to blockade the Argo in the \ vr "S Xi/x?)z> during the visit to Cyzicus 211
and were slain by Heracles and his comrades; but Deiloehus 212 ascribes the blockade to the
Pelasgians Kara t'xOos to 7Tp6s to'is 0err<iXoiy v(p' hv i^elieS'KrjVTO, 213 and says 214 that the
yrjyevets were Qeacra^ols (i.e. AoXiWi) eyxeipoydo-ropas, and that it was they who mistook the
Argonauts for pirates and planned the attack on them : an obvious attempt to relieve the
Thessalian Cyzicenes from the reproach of that blunder. Stephanus also (s.v. Be'aftiKos)
distinguishes the obstructive yrjyevees from the Pelasgians, but curiously reckons the
Pelasgians as allies of Heracles in his destruction of the yrjyevets.

Conon also adds that varepov (i.e. after the fight with Jason) the surviving Doliones 1770
Tvppr^vuiv Kv^ikov p(Tavi<Ttr](Tav <a\ Tvpprjvol ttjv ~K(pp()vqTov i'cr^ov : and that it was these
'Tyrrhenians' whom the Milesian colonists found there. Conon therefore had also before
him, besides the ' Pelasgic ' view of Ephorus, this other story which distinguished the
Doliones of Cyzicus from ' Tyrrhenian ' marauders in Hellespont; and we may well
believe that, writing as late as he did, he meant by ' Tyrrhenian' to signify much the same
as the H(~\a<TyiKov apea of Apollonius.

We reach therefore this conclusion. Attractive and accepted as it was, the Ephoran
view, that the Doliones were Pelasgians from Thessaly, did not wholly eclip>e an alternative
legend that they belonged to the same great Pagasaean adventure-cycle as the Argonauts
themselves ; and that in their Hellespontine home they and their friends were exposed to
the attack, not merely of half-conquered yqyevfes (avroxtioves) on their own side of the
water, but also of enemies from the European shore. These enemies Apollonius still calls
' Pelasgian : : only a later compiler like Conon uses the marine equivalent 'Tyrrhenian. 5

And this glimpse of another tradition does not stand quite alone. One of the theories,
we may remember, to account for the Dolopes of Scyros and the men of Sciathos, was that
they were EleXao-yoi e«c QpaK.rjs 21 -' 8ta/3dires, as Xoyoj : and we know that in the Homeric
Age there were already 'actual' Pelasgians as far afield as Crete. We must remember also
that Placie and Scylace, where Herodotus knew of Pelasgians surviving and speaking
'Pelasgic' in the fifth century, are in the immediate neighbourhood of Cyzicus itself.
Mela 210 moreover has a very similar suggestion about the Doliones themselves, for he
brings both them and their king Cyzicus not from Thessaly or Euboea, but from Thrace,
making them, in fact, almost an advanced guard of our immigrant Pelasgians from the
same region ; so that it is not impossible that here we may have a clue to the origin of
that ' Pelasgian ' ancestry or quality of the Doliones of C\ zicus, which attracted the
attention of Ephorus, and led to their incorporation in the great Pelasgian mythology.

It seems probable, then, that we may infer that what is present to the mind of

'■""■ Nctrr. n.

-"' Orpb. Arg. 497 "

■""• j.. 221 above.

• Scbol. Ap. bli. i. 921, 987.

210 Scbol. Ap. Mi. i. 1037.
" Ap. Rh. i. 987.
-'- Scbol. Ap. Rh. i. 1037.

- 1S Schol. Ap. Rh. i. 987.

- 14 Scbol. 989.

"■ ' Scymnus, f. S 4 , see p. '-'21 above.

216 Mela i. 19, 2. Compare Strabo's discus-
sion of the ethnography of all this region, sum-
marized on p. li'ti above.

A HISTORY OF THE PELASGIAN THEORY.

U -Jo

Apollonius anil some other late writers is a picture of a Thrace which the Hellespont,
as in Homer's time, tWos- e's'pyet with difficulty ; and of an Asiatic coast watched, like a
'Saxon shore,' day and night for the 'Winged Hats' from the European side.

But all this breathes a quite different atmosphere from that of the Alexandrian
Library. It presumes the existence of the Thraceward Pelasgians of the Catalogue, of whom
no single Greek writer. I think, takes any positive account till Strabo ; and even Herodotus
only implicitly and vaguely. It comes to us in a context,— the foundationdegend of
Cyzicus, and the ritual mWos of it.- slain founder-king,- which we can trace in nomenclature
back to Hecataeus of Miletus, and consequently beyond the period where the Ephoran theory
of a Pelasgian conquest begins to predominate over all : back, in fact, into days when
Lemnian Pelasgians were known to be post-Argonautic, and the Pelasgians of Placie and
Scylace were still talking their own language and recounting their own traditions.

It gives us, in fact, a very strong case for believing that here, at any rate, Apollonius
is incorporating, almost verbally, a section of a very much older Argonautica ; that this
Argonautica goes up certainly into the early days of Milesian colonization, probably into
the Homeric Age ; and very possibly even to a generation which stood to the Argonauts
and the Doliones as Demodocus stood to the Trojan War.

J. L. M.

H.S. — VOL. XXVII.

THREE INSCRIPTIONS FROM ASIA MINOR.

1. FRAGMENT of white marble, entire at the upper and right edge only,
measuring 5^ in. x 4 in. Found among the ruins of Troy on Apr. 20, 1907,
by Mr. F. G. Harman from the ' Argonaut ' : now the property of J. Alison
Glover, Esq., M.D.

«v,

Ai4>lAOSSAN°I

IlTANHrtPE^'
n* OEN TAy$EATX

•eaao

- - - " " o){v)

__------------ eypaxfrev - - ■

------------- AfofuXo? z^avdlir-

7rou - - - - €K - - t]j} iravrfyvpet, toj[v

X\ava6r}vala>v - - arecpaj^codeura vcp' ear5)\y o
---------------- ea \ - - -

The date is late, not earlier than the first century B.C., as is indicated by
the absence of the iota adscriptum (1. 4), and the coarse style of the lettering.
The form earcov in 1. 5 is characteristic of the first century B.C. 1 Am/>A.o9
"BavBliTTrov is not otherwise known to me. The irav^yvp^ or festival
assembly of the Panathenaea at Troy celebrated by the nine cities of the Ilian
union is mentioned in numerous inscriptions. 2 It is impossible to define
further the outline of the original document, as so much is lost.

2-3. 'Two tablets of marble, now in the house of M. Jean Gaetano, a silk-
manufacturer, at Moudania, on the sea of Marmora. Found by him whilst
diff°ine foundations for a wall near the centre of the town. 2, measuring

1 Meisterhans-Schwyzer, Grammatik der att. 461 ; Dittenberger, Syll. 2 503 ; Or. Or. Inscr.

Inschr.' p. 154 ; Dittenberger, Syll. 2 328, note Scl. 444, notes 1 anil 6. rw[v eWa iroXewv is

14_ also a possible restoration.

1 K.y. DSrpfeld, Troja ». Won, ii. p. 454,

THREE INSCRIPTIONS FROM ASIA MINOR. 227

about 3 ft. by 1 ft., was buried some six feet beneath the surface, and was
found in an upright position. Of 3 (which is smaller, about 9 in. wide and
6 in. high, and found near 2) nothing further is noted. Both tablets are
ornamented with serpents, 3 having a coiled serpent above in relief, 2 an
extended serpent incised. 2 is broken into several fragments ; but the surface
is wonderfully fresh (the lines ruled by the stone-cutter being still quite
obvious), I understand from M. Gaetauo that traces of an ancient channel or
waterway were found beside these stones.' (Note by J. A. Glover, Esq., M.D.)

ATAQHTYXH

ETTHKOHOEn

ACKAHnin

EniAAYPin

TTEPrAMHNn 5

AinPYTAKATOIKoYNTI
rNAIBIOCIOYCToC

YrrEPcnTHPiAC

THNCTOANANE

CTHCENKAITON 10

EICTHNOIKOAO

MIANTOnON

EXAPICATO

THGEn

'Ayadfj ti>XV 'E'Ttijkoco dew XaK\i]7riM 'E,7ri8avp{'p Uepya/iyjva) Sicopvya
/caToiKOvvTi T. Nat'/3i09 'Iovcttos inrep acoTtjpi'at ri]v crroav kclI tov et? tijv
oiKoho/iiav roirov £x a pL aaT0 T( f> &€<*>•

TPAKXIIIEKOYNAOI
OP(j) ANOIYn E P I A I
A I IHT H P I A I AhE
OHCAAIKAHni-Q
AIOPYTEITH 5

VpaKXiS 2.€/covi'&o<; 6p(j>avo<; vnrep i&la*; aa)ri]pc'a^ aviOi-jKa 'Actk\j]771(i)
Ai[co]pvyeiry].

On April 19, 1907 ;i number of us who were travelling on board the
' Argonaut * landed at Moudania. and thouce proceeded by rail to Brusa in
Bithynia. While in Moudania our doctor. Mr. J. Alison Glover. M.D., heard
of these inscriptions, and went to see them. He took copies and memoranda

Q 2

228 THREE INSCRIPTIONS FROM ASIA MINOR.

of both : of 3 lie made also a rough rubbing. These he has handed to me
for publication. They are not in Boeckh's Corpus, and I do not remember to
have seen them before. The inscriptions are entire and legible. The last
line of 3 is slightly obscured by the carved margin of the stone ; but no
doubt the third latter is n, and not 0- Dr. Glover writes c carefully in 2 :
his rubbing and copy of 3 give 2. Iota adscript is absent.

Myrlea, renamed by Prusias Apameia after his wife, and now known as
Moudania, was in ancient days, as now, a natural landing place for those who
wanted to reach Prusa. The distance between the two towns as the crow flies
is about ten miles : by rail or road, nearer twenty. Prusa itself is noted for its
water-supply, and its baths were as famous in antiquity (Pliny, Ep. to Tr.
xxiii) as to-day. Of the social history of Myrlea we know practically nothing.
The inscriptions before us belong to the second century a.d. They indicate
that there was an altar or temple at Myrlea to Asklepios, and that near his
sanctuary, or through it, there flowed a stream of pure water artificially
brought by a conduit (Stwpvf;), possibly from the medicinal stream running
down from the springs at Brusa.

In 2 Asklepios is spoken of as the god of Epidauros and of Pergamon.
Readers of Aristeides the Orator will fully appreciate the epithet Uepyafxrjvo}?
A Roman named Caius Naevius Justus makes a free sift to the sod of a
' stoa,' or colonnade, and the site on which he built it. This dedication he
makes inrep acortjpta^, i.e. probably he had escaped harm in an epidemic,
and had attributed his immunity to the favour of the god. 'Ew^/com
suggests that it was in answer to prayer. 1 It may be noted that Sicopvya
(line 6) is regarded as a less correct form than hiajpv%a, while olKohoplav is
certainly better Attic than oUoSo/uijv (cp. van Herwerden, Lex. s.v.).

3, which cannot be much later, and is perhaps earlier, than 2, is a
dedication to the same deity, here called AicopvyecTi]*;, by one Gracchis
Secundus, vTrep ISicis aa>Tr)pLas. Probably he had lost one or both of his
parents in an epidemic (dpcpavos), and attributes his own safety to the favour
of the god. The name Gracchis is unique, so far as I know. Happily the
evidence of the marble is beyond question.

The Roman names remind us that Apameia (Myrlea) was a colonia: see
Hardy's note on Apameia in his edition of Pliny's CorresjJondcncc with Trajan,
p. 148, Ep. xlvii.

E. L. Hicks.

- Cp. I'aus. ii. 26. 8 on the relation between and the Peiraeus inscr. AeA-r. 1S8S. p. 134,
the cult.- :it Pergamon and Epidaurus. No. 20.

4 Op. also Fouillcs d'JJlpid. p. 58, No. 11"

J. H. S. VOL. XXVII. (1907). PL. XVI.

-I V

o . :

u. ^
O .§

UJ t.

w -

< §>

CO i

I- §
<

z
cc

LlI

CO
<
>

z
o

J. H. S. VOL. XXVII. (1907) PL. XV.

MONEMVASIA. A. — Havajia MvpTiSiwrtacra.

{Photograph by Mr. A. J. B. Wace.

MONEMVASIA. B.— Ayia So^ta.

{Photograph by Mr. A. J. B. Wace.

MONEMVASIA.
[Plates XV., XVI.]

MONEMVASIA DURING THE FlIAXKISH PERIOD (1204-1540).

There are few places in Greece which possess the combined charms of
natural beauty and of historic association to the same extent as Monemvasia,
The great rock which rises out of the sea near the ancient Epidauros Limera
is not only one of the most picturesque sites of the Peloponnese, but has a
splendid record of heroic independence, which entitles it to a high place in the
list of the world's fortresses (Figs. 1, 2). Monernvasia's importance is, however,
wholly mediaeval ; and its history has hitherto never been written ; for the
painstaking brochure of the patriotic Monemvasiote deputy and ex-Minister
K. Papamichalopoulos, was composed before modern research rendered it
possible to draw upon the original authorities at Venice and elsewhere. In
the present paper I have endeavoured to state briefly what, in the present
state of Greek mediaeval studies, is known about this interesting city during
the Frankish period.

Fig. 1. —Monemvasia from riiE Land.
{Photograph by Mrs. Miller.)

At the time of the Frankish Conquest of the rest of Greece, Monemvasia
was already a place of considerable importance. Even if we reject the state-
ment of the fifteenth century historian, Phrantzes, 1 himself a native of the
place, that the Emperor Maurice had raised it to the rank of the 34th Metro-
politan see — a statement contradicted by an ecclesiastical document of 1307
— we know at least that it was even then the seat of a Greek bishopric,
whose holder remained a suffragan of Corinth - till the Latins captured the

1'. 398. 37 .;• <■ , ii. 287 ; Dorotfieos of

Miklosich und Miiller, Acta et I ". isia, B./J.uV laropiKov (ed. 1814), 397.

230

WILLIAM MILLER

latter city in 1210, The Comneni had confirmed the liberties of a community
so favourably situated, and the local aristocracy of Monemvasia enjoyed the
privilege of self-government. Thanks to the public spirit of its inhabitants,
the wisdom of the local magnates, and the strength of its natural defences,
which made it in the Middle Ages the Gibraltar of Greece, it had repelled
the attack of the Normans from Sicily in the middle of the twelfth century.
Fifty years later it was a busy sea-port town, whose ships were seen at the

Fig. 2. — Monemvasia. Entrance to Kastro.
(Photograph by Mr. A. J. B. Waee.)

Piraeus by Michael Akominatos, the last Metropolitan of Athens before the
Conquest, and whose great artistic treasure, thc^ famous picture of Our Lord
being 'dragged,' which has given its name to the 'JL\ko/j,€vo<; church, attracted
the covetousness of the Emperor Isaac II. 3

As might have been expected from its position and history, Monemvasia
was the last spot in the Peloponnese to acknowledge the Frankish supremacy.
Geoffrey I. Villehardouin had contented himself perforce with sending a body

Lampros, Mix^tjA 'Ako/j.lvo.tov, ii. 137 ; Niketas, 97, 581-92.

MONEMVASIA. 231

of troops to raid the country as far as the causeway, or /xovr) efjb/3acn<i,
which leads to the great rock-fortress and from which its name is
derived; 4 and his son Geoffrey II. seems to have meditated the con-
quest of the place ; 5 but it was reserved for the third of the Villehardouins,
soldierly Prince William, to hoist the croix ancrdc of his family over the
'sacred rock' of Hellenism, which was in uninterrupted communication by
sea with the successor of Byzantium, the Greek Emperor of Nicaea, 6 and was
therefore a constant source of annoyance to the Franks of the Peloponnese.
The Prince, after elaborate preparations, began the siege in 1245. He sum-
moned to his aid the great vassals of the Principality — Guy I. of Athens, who
owed him allegiance for Nauplia and Argos ; the three barons of Euboea ;
Angelo Sanudo, Duke of Naxos, with the other lords of the Cyclades, and the
veteran Count Palatine of Cephalonia, Matteo Orsini. ruler of the island-
realm of Odysseus. 7 But the Prince of Achaia saw that without the naval
assistance of Venice, which had taken care that his principality should not
become a sea-power, he could never capture the place. He accordingly
obtained the aid of four Venetian galleys, and then proceeded to invest the
great rock-fortress by land and water. For three long years the garrison held
out, ' like a nightingale in its cage,' as the Chronicler quaintly says — and the
simile is most appropriate, for the place abounds with those songsters — till
all supplies were exhausted, and they had eaten the very cats and mice.
Even then, however, they only surrendered on condition that they should be
excused from all feudal services, except at sea, and should even in that case be
paid. True to the conciliatory policy of his family, William wisely granted
their terms, and then the three arclions of Monemvasia, Mamonas, Daimon-
oyannes, and Sophianos, advanced along the narrow causeway to his camp and
offered him the keys of their town. The conqueror received them with the
respect of one brave man for another, loaded them with costly gifts, and gave
them fiefs at Vatika near Cape Malea. A Frankish garrison was installed in
the coveted fortress ; and a Latin bishop, Oddo of Verdun, at last occupied
the episcopal palace there, which had been his (on paper) ever since
Innocent III. 8 had organised the Latin see of Monemvasia as one of the
suffragans of Corinth.

The Frankish occupation lasted, however, only fourteen years, and has left
no marks on the picturesque town. Buchon, indeed, who spied the Ville-
hardouin arms on the Gorgoepekoos church at Athens, thought that he had
discovered the famous croix ancrec on oue of the churches. 9 He apparently
meant the 'EX/to/ievo? Church, which the late Sir T. Wyse called and
Murray's Handbook still calls St. Peter's— a name not now known in Monem-
vasia, but derived perhaps from an inscription to a certain Dorninus Petnis,

4 Tb XpoviKov rov Mopeais, 1. 2065. contingents.

" Ibid. 11. 2630, 2644. B Epislolac, vol. ii. p. 622 ; Les Begistres

6 Ibid. 11. 2765-9. d'Lmoccnt IT. vol. iii. 306. 397.

7 Ibid. 11. 2891-6 ; Romanos, rpanavbs ' La Grece Conlincntah, p. 412 ; Sir T.
ZdpCv^. 136. The French version of the Wyse, Excursion Into the Peloponnesus, i. 6.
'Chronicle' omits the Naxian ami Cephalonian Cf. Mr. Tozer in J.H.S. iv. 233-6.

232 WILLIAM MILLER

whose remains 'lie in peace' hard by. One church in the town, 'Our Lady
of the Myrtle,' bears, it is true, a cross with anchored work below, and four
stars above the door. But this church, as I was informed and as the name
implies, was founded by people from Cerigo, whose patron saint is the Havayia
MvpTiSicoTiaaa (Pi. XV. A). The capture of the town by the Franks is,
however, still remembered at Monemvasia, and local tradition points out the
place on the mainland where Villehardouin left his cavalry. One pathetic
event occurred at the rock during the brief Frankish period — the visit of the
last Latin Emperor of Constantinople, Baldwin II., in 1261, on his way from
his lost capital to Italy. 10 In the following year Monemvasia was one of the
castles ceded to his successor, the Emperor Michael VIII. Palaiologos, as the
ransom of Prince William of Achaia, captured by the Greeks three years
earlier after the fatal battle of Pelagonia.

The mediaeval importance of Monemvasia really dates from this retro-
cession to the Byzantine Emperor in 1262, when a Byzantine province was
established in the south-east of the Morea. It not only became the seat of an
Imperial governor, or tcecfraX)], but it was the landing-place where the Imperial
troops were disembarked for operations against the Franks, the port where
the Tzakones and the Gasmouloi, or half-castes, of the Peloponnese enlisted
for service in the Greek navy. During the war which began in 1263 between
Michael VIII. and his late captive, we accordingly frequently find it men-
tioned ; it was thither that the Genoese transports in the Imperial service
conveyed the Greek troops ; it was thither, too, that the news of the first
breach of the peace was carried post-haste, and thence communicated to
Constantinople ; it was there that the Imperial generals took up their head-
quarters at the outset of the campaign ; and it was upon the Monemvasiotes
that the combatants, when they were reconciled, agreed to lay the blame for
the war. 11 Under the shadow of the Greek flag, Monemvasia became, too,
one of the most dangerous lairs of corsairs in the Levant. The great local
families did not disdain to enter the profession, and Ave read of both the Daimon-
oyannai and the Mamonades in the report of the Venetian judges, who drew
up a long statement in 1278 of the depredations caused by pirates to Venetian
commerce in the Levant. On one occasion the citizens looked calmly on
while a flagrant act of piracy was being committed in their harbour, which, as
the port of shipment for Malmsey wine, attracted corsairs who were also
connoisseurs. 1 ' 2 Moreover, the Greek occupation of so important a position
was fatal to the Venetian lords of the neiohbourinof islands, no less than to
Venetian trade in the Aegean. The chief sufferers were the two Marquesses
of Cerigo and Cerigotto, members of the great families of Venier and Viaro,
who had occupied those islands after the Fourth Crusade. It would appear
from a confused passage of the Italian Memoir on Cerigo, that the islanders,

Tb Xpovncov tov Mo/>e'a>?, 1. 1306; /.- Livrc 4643, 502(5, 5509, 5576.

de la Conqucste, p. 27. 12 Fontcs Rcrum Austriacarum, Abt.

" Lea Rcgistresd'Urbain II'. ii. 100, :J41 ; Tb 15. xiv. 164, 192 :;, 204, 215, 220, 226, 248.
Xpoumbv tov Moptus, II. 1504, 4547,4580,4534,

MONEMVASIA. 233

impatient at the treatment which they received from their Latin lord, the
descendant, as he boasted, of the island-goddess Venus herself, sent a depu-
tation to invoke the aid of the Greek governor of the new Byzantine province
in the Morea. 18 At any rate, the famous cruise of Licario, the upstart Italian
of Negroponte who went over to the Greeks, temporarily ended the rule of
the Venetian Marquesses. A governor was sent to Cerigo from Monemvasia ;
but ere long Michael VIII. conferred that island upon the eminent Monem-
vasiote archon, Paul Monoyannes, who is described in a Venetian document as
being in 1275 'the vassal of the Emperor and captain of Cerigo.' Mono-
yannes fortified the island, where his tomb was discovered during the British
protectorate, and it remained in the possession of his family till 1309, when
intermarriage between the children of its Greek and Latin lords restored
Cerigo to the Venieri. 14

The Byzantine Emperors naturally rewarded a community so useful to
them as that of Monemvasia. Michael VIII. granted its citizens valuable
fiscal exemptions; his pious son and successor, Andionikos II. not only con-
firmed their privileges and possessions, but founded the church of the Divine
Wisdom which still stands in the castle. The adjoining cloister has fallen in
ruins; the Turks after 1540 converted the church, like the more famous Santa
Sophia of Constantinople, into a mosque, the mihrab of which may still be
traced, and smashed all the heads of the saints which once adorned the church
—an edifice reckoned as ancient even in the days of the Venetian occupation,
when a Monemvasiote family had the /its patronatm over it (PI. XV. B). But
a fine Byzantine plaque over the door— two peacocks and two lambs— still
preserves the memory of the Byzantine connexion. Of Andionikos II. we
have, too, another Monemvasiote memorial— the Golden Bull of 1293, by
which he gave to the Metropolitan the title of ' Exarch of all the Pelopon-
nesos/ with jurisdiction over eight bishoprics, some, it is true, still in partibus
infidelium, as well as the titular Metropolitan throne of Side, and confirmed
all the rights and property of his diocese, which was raised to be the tenth of
the Empire and extended, at any rate on paper, right across the peninsula to
' Pylos, which is called A varinos ' — a convincing proof of the error made by
Hopf in supposing that the name of Navarino arose from the Navarrese
company a century later. The Emperor lauds in this interesting document,
which bears his portrait and is still preserved in the National Library and (in
a copy) in the Christian Archaeological Museum at Athens, the convenience
and safe situation of the town, the number of its inhabitants, their affluence
and their technical skill, their seafaring qualities, and their devotion to his
throne and person. His grandson and namesake, Andronikos III. in L332,
granted them freedom from market-dues at the Peloponnesian fairs." But a
city so prosperous was sure to attract the covetous glances o( enemies.

is Antique Mcmoru di Cerigo, apud Sathas, Cronol ■ 1 '" 1 - w ' : " ;

Mi/iju«a 'EAA»jnK7}s 'Iffroplas, vi. 301. Krscli und Grubcr, lxxxv. 310. _

» Sauudo, Istoria del Regno, apud Hopf, l5 Miklosich and MulU-r. op. at v. loo ol ;

Chroniques Grteo-r us, 127 ; Fontes lierum PhrantzSs, 399, 100 : Dorotl sol Monemvasia,

Austriacarinn, Abt. ii. B. xiv. 181 ; Sansovino, B^AiV '\<jto P lk6v. ■

231 WILLIAM MILLER

Accordingly, in 1292, Roger de Lluria, the famous admiral of King James of
Aragon, on the excuse that the Emperor had failed to pay the subsidy
promised by his father to the late King Peter, descended upon Monemvasia,
and sacked the lower town without a blow. The archons and the people took
refuge in the impregnable citadel, leaving their property and their Metro-
politan in the power of the enemy. 10 Ten years later, another Roger, Roger
de Flor, the leader of the Catalan Grand Company, put into Monemvasia on
his way to the East on that memorable expedition which w r as destined to ruin
' the pleasaunce of the Latins ' in the Levant. On this occasion the Catalans
were naturally on their good behaviour. Monemvasia belonged to their
new employer, the Emperor Andronikos ; it had been stipulated that they
should receive the first instalment of their pay there ; and Muntaner 17 tells
us that the Imperial authorities gave them a courteous reception and pro-
vided them with refreshments, including probably a few barrels of the famous
Malmsey.

Monemvasia fortunately escaped the results of the Catalan expedition,
which proved so fatal to the Duchy of Athens and profoundly affected the
North and West of the Morea. Indeed, in the early part of the fourteenth
century the corsairs of the great rock seemed to have actually seized the
classic island of Salamis under the eyes of the Catalan rulers of Athens,
whose naval forces in the Saronic Gulf had been purposely crippled by the
jealous Venetian Government. At any rate we find Salamis, which had
previously belonged to Bonifacio da Verona, the baron of Karystos in Euboea,
and had passed with the hand of his daughter and heiress to Alfonso
Fadrique, the head of the terrible Catalan Company in Attica, now paying
tribute to the Byzantine governor of Monemvasia. 18 When, however,
towards the end of the fourteenth century, the Greeks began to recover most
of the Peloponnese, the city which had been so valuable to them in the earlier
days of the reconquest of the Morea had to compete with formidable rivals.
In L897, when Theodore I. Palaiologos obtained, after a desperate struggle,
the great fortress of Corinth, which had been his wife's dowry from her
father, Xerio Acciajuoli, his first act was to restore the Metropolitan see of
that ancient city, and the first demand of the restored Metropolitan was for
the restitution to him by his brother of Monemvasia of the two suffragan
bishoprics of Zemenos and Maina, which had been given to the latter's
predecessor after the Latin conquest of Corinth. 19 This demand was granted,
and we are not surprised to hear that the Monemvasiotes were disaffected
to the Despot, under whom such a slight had been cast upon their Church.
The Moreat archons at this period were intensely independent of the
Despot of Mistra, even though the latter was the brother of the Emperor.
The most unruly of them all was Paul Mamonas of Monemvasia, who

16 Lc Lucre de la Conqueste, 363 ; Libro de los [7 ('lis. 199, 201.

/' ios, 107 ; Muntaner, Cronaca, ch. 117 ; I8 Thomas, Diplomatarium Veneto-Levan-

Burtholomaeus de Neocastro and Nieolaus tinum, i. 127.

Specialis apud Muratori Iter. Ital. Script, xiii. '•' Miklosich und Miiller, I.e.

11-." ; :;. 959.

MONEMVASIA. 235

belonged to the great local family which had been to the fore in the days of
Villehardouin. This man held the office of 'Grand-Duke' or Lord High
Admiral in the Byzantine hierarchy of officials and claimed the hereditaiy
right to rule as an independent princelet over his native city, of which his
father had been Imperial governor. When Theodore asserted his authority, and
expelled the haughty archon, the latter did not hesitate to arraign him before
the supreme authority of those degenerate days — the Sultan Bajazet I. who
ordered his immediate restoration by Turkish troops — a humiliation alike
for the Greek Despot and for the sacred city of Hellenism. 2 ' Theodore had.
indeed, at one time thought of bestowing so unruly a community upon a
Venetian of tried merit ; and, in 141!), after the death of Paul's son, the
Republic appears actually to have come into possession of the coveted rock
and its surroundings — then a valuable commercial asset because of the
Malmsey which was still produced there.- 1

It was at this period that Monemvasia produced two men of letters,
George Phrantzes and the Monk Isidore. To the latter we owe a series of
letters, one of which, addressed to the Emperor Manuel II. on the occasion of
his famous visit to the Morea in 1415, describes his pacification of Maina and
his abolition of the barbarous custom of cutting off the ringers and toes of the
slain, which the Mainates had inherited from the Greeks of Aeschylus and
Sophocles. He also alludes to the Greek inscriptions which he saw at
Yitylo.-- Of Phrantzes, the historian of the Turkish conquest, the secretary
and confidant of the Palaiologoi, the clever if somewhat unscrupulous
diplomatist, who, after a busy life, lies buried in the quiet church of SS.
Jason and Sosipater at Corfu, it is needless to speak. In the opinion of the
writer, Phrantzes should hold a high place in Byzantine history. His style
is clear and simple, compared with that of his contemporary Chalkokondyles,
the ornate Herodotus of the new Persian Conquest ; he knew men and
things; he was no mere theologian or rhetorician, but a man of affairs : and
he wrote with a na'in/r, which is as amusing as it is surprising in one of his
profession. Monemvasia maybe proud of having produced such a man, who
has placed in his history a glowing account of his birthplace. We hear too
in 1540 of a certain George, called 'Count of Corinth' but a uative of
Monemvasia, who had a fine library, and among the many Peloponnesian
calligraphists, the so-called ' Murmures,' found later on in Italy there were
some Monemvasiotes.-*

The Venetians did not, on this occasion, long retain Monemvasia. A
few years later we find it in the possession of the Despot Theodore II.
Palaiologos,' 24 who ratified its ancient privileges. All the Despots subjects,
whether freemen or serfs, were permitted to enter or leave this important

-' Phrantzes, ;"7 : Manuel Palaiologos, Theo- '-"-' Ne'us 'EWrjioui-iiuccv, i. 269; ii. 181.

dori Despoti Laudatio Funebris, apud Migne, : Montfaucon, Palaeographia Gr

Patroloyia Gfraeca, clvi. 228 9 ; Chalkokondyles, 'E\A7jro / ur7)jU">»'> 336-46.
SO. M Miklosich and Muller, v. 171-1 ; napia*-

21 Misti, liii. fol. 44, 'J56 ; hi. fol. 76 v., <ros, vii. A72-0.
cited by Hopf, cp. dt. lxxxvi, 79.

236 WILLIAM MILLER

city without let or hindrance, except only the dangerous denizens of
Tzakonia and Vatika, whose character had not altered in the two hundred years
which had elapsed since the time of Villehardouin. The citizens, their
beasts, and their ships were exempt from forced labour; and, at their special
request, the Despot confirmed the local custom, by which all the property of
a Monemvasiote who died without relatives was devoted to the repair of the
castle; while, if he had only distant relatives, one-third of his estate was reserved
for that purpose (PI. XVI. A). This system of death duties (to ciQuotiklov, as it
was called) was continued by Theodore's brother and successor, Demetrios,
by whom Monemvasia was described as ' one of the most useful cities under
my rule.' 25 Such, indeed, he found it to be, when, in 1458, Mohammed II.
made his first punitive expedition into the Morea. On the approach of the
great Sultan, the Despot fled to the rock of Monemvasia. It was the
ardent desire of the Conqueror to capture that famous fortress, ' the
strongest of all cities that we know,' as the contemporary Athenian
historian, Chalkokondyles,-' 1 called it. But his advisers represented to him
the difficult nature of the country which he would have to traverse, so he
prudently desisted from the enterprise. Two years later, when Mohammed
II. visited the Morea a second time and finally destroyed Greek rule in that
peninsula, Monemvasia again held out successfully. After sheltering Deme-
trios against an attack from his treacherous brother Thomas, the town gave
refuge to the wife and daughter of the former. Demetrios had, however,
promised to give his daughter in marriage to the great Sultan ; and Isa, son
of the Pasha of Uskub, and Matthew Asan, the Despot's brother-in-law, were
accordingly sent to demand the surrender of the city and of the two
princesses, whom it contained. The Monemvasiotes did, indeed, hand over
the two Imperial ladies to the envoys of the Sultan and the Despot; but,
relying on their immense natural defences, animated by the sturdy spirit
of independence which had so long distinguished them, and inspired by the
example of their governor, Manuel Palaiologos, they bade them tell
Mohammed not to lay sacrilegious hands on a city which God had meant
to be invincible. The Sultan is reported to have admired their courage, and
wisely refrained from attacking the impregnable fortress of mediaeval
Hellenism. As Demetrios was the prisoner of the Sultan, the Governor
proclaimed Thomas as his liege jlord ; but the latter, a fugitive from Greece,
was incapable of maintaining his sovereignty and tried to exchange it with
the Sultan for another sea-side place. 27 A passing Catalan corsair, one
Lope de Baldaja, was then invited to occupy the rock ; but the liberty-loving
inhabitants soon drove out the petty tyrant whom they had summoned to
their aid, and, with the consent of Thomas, placed their city under the
protection of his patron, the Pope. Pius II. gladly appointed both spiritual
and temporal governors of the fortress which had so Ions been the stronghold

Ibid. iii. 258. -'" Chalkokondyles, 476,485; Phrantzes, 396-

P- ll7 - 97; Spandugino (ed. 1551), 14 5.

MONEMVASIA. 237

of Orthodoxy, and of that nationalism with which Orthodoxy was

identical. 28

But the papal flag did nut wave long over Monemvasia. The Orthodox
Greeks soon grew tired of forming part of the Pope's temporal dominion, and
preferred the rule of Venice, the strongest maritime power interested in the
Levant, whose governors were well known to be ' first Venetians and then
Catholics.' The outbreak of the Turco-Venetian War of 1403, and the
appearance of a Venetian fleet in the Aegean, gave the citizens their oppor-
tunity. The Pope, as Phrantzes informs us, had no wish to give up the
place ; but he was far away, his representative was feeble, the flag of Venice
was for the moment triumphant in Greek waters, and accordingly in 14G3 or
1464, the inhabitants admitted a Venetian 'garrison. On September 21,
1404, the Senate made provision for the government of this new dependency.
A Podestd was to be elected for two years at an annual salary of 500 gold
ducats, this salary to be paid every three months out of the revenues of the
newly-conquered island of Lemnos.«- Six months later, it was decreed that
in case there was no money available for the purpose- at Lemnos, the Podestd
should receive his salary from the Cretan treasury.-" From that time to
1540 Monemvasia remained a Venetian colony. Once, indeed, a plot was
organised in the ancient city of the Palaiologoi for the purpose of wresting
the place from the claws of the Lion of St. Mark. Andrew Palaiologos, the
still more degenerate son of the degenerate Thomas, had, in 1494, transferred
all his Imperial rights and claims to King Charles VIII. of France, then
engaged in his expedition to Naples, in the Church of San Pietro in Montorio
at Rome. In accordance with this futile arrangement, his partisans at
Monemvasia, where the Imperial name of Palaiologos was still popular,
schemed to deliver the city to his French ally. 30 But the plans of Charles
VIII., and with them the plot at Monemvasia, came to nought. Venice
remained mistress of the Virgin fortress.

Down to the peace of 1502-3, Monemvasia seems to have been fairly
prosperous under Venetian rule. By the Turco-Venetian treaty of 1479 she
had been allowed to retain the dependency of Vatika :;1 in the neighbourhood
of Cape Malea, which had been captured from the Turks in 1463, and whore
her citizens had long possessed property. But the territories of Monemvasia
were terribly restricted after the next Turco-Venetian war: she had then
lost her outlying castles of Rampano and Vatika, from which the ecclesi-
astical authorities derived much of their dues ; and we find the inhabitants
petitioning the Republic for the redress of their grievances, and pointing out

28 Magno, Annul; Veneti, apud Hopf, I'hron- 1464. ami the Venetian documenl above-quoted
iqucs Greco-romanes, 203-4; Pii II. Com- points to that year ; but Malatesta's secretary
mcntari 103-104. i" his account of the war (Sathas, I.e.) puts it

29 PhrantzSs, 11.". ; Magno, 204 ; Sathas, vi. in 1463, before tin- siege of Corinth.
95 ; Chalkokondyies, 556. K gina, fol. 52, 56 Sanudo, Diarii, i. 703.

(for a copy of which I am indebted to Mr. :l Predelli, Commemoriali, v. 228

Horatio F. Brown : see Appendix), The actual 241 : Miklosich u. Muller, op. cit. iii. 2>'3-309.

date is uncertain ; PhrantzSs and Magno give

238

WILLIAM MILLER

that this last delimitation of their frontiers had deprived them of the lands
which they had been wont to sow. The rock itself produced nothing, and
accordingly all their supplies of corn had now to be imported through the
Turkish possessions. 32 As for the famous vintage, which had been the delight
of Western connoisseurs, it was no longer produced at Malvasia, for the Turks
did not cultivate the vineyards which were now in their hands, and most of
the so-called ' Malmsey,' nihil de Malfasia habens scd nomen, as worthy Father
Faber says, had for some time come from Crete or Modon, 33 till the latter
place, too, became Turkish. But, in spite of these losses, Monemvasia still
remained what she had been for centuries — an impregnable fortress, the

Fig. 3. — Kastro.
(Photograph by Mr. A. J. B. Wacc.)

Gibraltar of Greece. The Venetians renewed the system, which had prevailed
under the Despots of the Morea, of devoting one of the local imposts to the
repair of the walls ; the Venetian Podesta, who lived, like the military
governor, up in the castle, seems to have been a popular official ; and the
Republic had wisely confirmed the special privileges granted by the Byzantine
Emperors to the Church and Community of this favoured city (Fig. 3). Both
a Greek Metropolitan and a Latin Archbishop continued to take their titles
from Monemvasia, and the most famous of these prelates was the eminent
Greek scholar, Marcus Mousouros. It is interesting to note that in 1521

32 Sathas, M^TjiUeTa 'EWrjviKrjs 'laropias, iv.
230; Sanudo, iJiurii, xxix. 482.

a3 Feyerabend, Beyssbuch des Heyligens Lands,

fol. 182; Faber, Evagatorium, iii. 314. The

name was so long preserved that a wine-shop in
Venetian dialect was called 'Malvasia.'

MONEMVASIA 239

Pope Leo X. had a scheme for founding an academy for the study of the
Greek language out of the revenues of whichever of these sees first fell
vacant, as Arsenios Apostoles, at that time Metropolitan, was a learned
Greek and a Uniate, and in both capacities a prime favourite of the classi-
cally cultured Pontiff. In 1524, however, despite the thunders of the
Oecumenical Patriarch, the Greek and the Italian prelates agreed among
themselves that the former should retain the see of Monemvasia and that
the latter should take a Cretan diocese. 34 The connexion between 'the great
Greek island ' and this rocky peninsula was now close. The Greek priests of
Crete, who had formerly gone to the Venetian colonies of Modon and Coron
for consecration, after the loss of those colonies in 1500 came to Monemvasia;
the Cretan exchequer continued to contribute to the expenses of the latter ;
and judicial appeals from the Podesid of Malmsey lay to the colonial author-
ities at Candia, instead of being remitted to Venice ; for, as a Monemvasiote
deputation once plaintively said, the expenses of the long journey had been
defrayed by pawning the chalices of the churches. Even now Monemvasia
is remote from the world ; in those Venetian days she was seldom visited,
not only because of her situation, but because of the fear which ships
captains had of her inhabitants. 35

The humiliating peace of 1540, which closed the Turco-Venetian war of
1537, closed also the history of Venice in the Morea till the brief revival
at the close of the seventeenth century. This shameful treat} 7 cost the
Republic her two last possessions on the mainland of Greece — Nauplia
and Monemvasia, both still uncaptured and the latter scarcely assailed
by the Turkish forces. 30 Admiral Mocenigo was sent to break as best he
could to her loyal subjects the sad news that the Republic had abandoned
their homes to the Turks. The Venetian envoy, if we may believe the speech
which Paruta puts into his mouth, repeated to the weeping people the ancient
adage, ubi bene, ibi patria, and pointed out to them that they would be better
off in a new abode less exposed than their native cities had been to the Turkish
peril. In November a Venetian fleet arrived in the beautiful bay of Nauplia
and off the sacred rock of Monemvasia to remove the soldiers, the artillery,
and all the inhabitants who wished to live under Venetian rule. Then
the banner of the Evangelist was lowered, the keys of the two last Venetian
fortresses in the Morea were handed to Kassim Pasha, and the receipts for
their transfer were sent to Venice. 37

The inhabitants of the two cities had been loyal to Venice, and Venice
was loyal to them. The first idea of transporting the Mnnemvasiotes to the
rocky island of Cerigo — then partly a Venetian colon}" and partly under the
rule of the great Venetian family of Venier, which boasted its descent
from Venus, the fabled goddess of Kythera — was abandoned, in deference

34 Sanudo, Diarii, vii. 714 ; xxiii. 536 ; xxiv. Satlias, i\\ 2-24, 227, 229, 234 ; Lamansky, /.■ s

669; xxv. 64 ; xxix. 402; xxxi. 227 ; xxxv. Secrets dc F Elat Jc I'cnise. [>. 059; Feyerabend,

363; xliv. 475 ; lv. 296 ; Neos 'F.\\r)vonvr)txa>v, op. cit. fol. 112.

iii. 56. 3,i l'redelli, Commcmoriali, vi. 236, 2 -

33 Sanudo, Diarii, xi. 349; xxxiii. 360 > ;r Paruta, Historia Venetiana, L 451-3.

240

MONEMVASIA.

to the eloquent protests of the Metropolitan, and lands were assigned
to the exiles in the more fertile colonies of the Republic. A commission
of five nobles was appointed to consider the claims, and provide for the
settlement, of the stradioti, or light horsemen from Nauplia and Monem-
vasia, who had fought like heroes against the Turks ; and this commission
sat for several years, for the claimants were numerous and not all genuine. 38
Some, like the ancient local family of Daimonoyannes, formerly lords of
Cerigo, received lands in Crete, 39 where various members of the Athenian
branch of the great Florentine family of the Medici, which had been settled
for two hundred years at Nauplia, also found a home. Others were removed
to Corfu, where they soon formed an integral part of the Corfiote population
and where the name of these Stradioti is still preserved in a locality
of the island ; while others again were transplanted to Cephalonia, Cyprus, or
Dalmatia. Not a few of them were soon, however, smitten with home-
sickness; they sold their new lands and returned to be Turkish subjects at
Nauplia and Monemvasia. 40

The Venetian fortifications ; the old Venetian pictures on the eikonostasis
of the 'E\k6/j.€vo<; church ; the quaint Italian chimneys, and the well-head up

Fig. 4. — Arms on Well-Head in the Castle.
(Sketch by Mrs. Miller.)

in the castle, which bears the winged lion of St. Mark, two private coats of
arms, the date MDXIV and the initials S R upon it, the latter those of Sebas-
tiano Rnnier, Podcstd from 1510 to 1512, still speak to us of this first Venetian
occupation, when the ancient Byzantine city, after the brief vicissitudes
of French and Papal government, found shelter for nearly eighty years
beneath the flag of the Evangelist (PL XVI. B and Fig. 4).

William Miller.

:;s Lami, Dcliciac Eriiditorum, xv. 203 ;
Sathas. op. cit. viii. 310 3, 320 1, 335, 344,
377 8, 441-3.

'-'■' Ibid. 312, 413. 450, 454.
40 Ibid. 396.

MoXEMVASIA. 241

APPENDIX.
Two Venetian Documents relating to the Acquisition of

MONEMVASIA I .V 1464.

I. — Regina fol. •">:!.

mcccclxiiij indictione xij.

I >ie xxi Septembris.

Cum per gratiam omnipotentis Dei acquista sit in partibus grecie insula Staliminis
dives et opulenta in qua sunt tres terre cum Castellis viz Cochinum, Mudrum et Paleo-

rum que tempore pacis reddere solent ducatos circa x m . Item etiam Cdvitas Malvasie
sita in Amorea. Ad quorum locorum bonam gubernationem et conservationem sub
obedientia nostri Dominii providendum est de rectoribus et camerariis e venetiis mittendis
tarn pro populis regendis et jure reddendo quam pro introitibus earum bene gulieinandis et
non perdendis sicnt Imeusque dicitur esse factum ....

Eligatur per quattuor manus electionum in maioii consilio anus potestas Malvasie cum
salario ducatorum V. auri in anno, sit per duos annos tantum ; et habeat salarium liberum
cum prerogativis et exemptionibus rectoiis Staliminis et similiter in contumacia sua.
Debeat habere duos faniulcs et tres equos et recipiat salarium suum ab insula Staliminis de
tribus mensibus in tres menses ante tempus.

fDe parte 474

De non 14

Non syncere ..... 9

Die xvij Septembris mcccclxiiij in conselio di xl ta .

De parte . . . . . 26

De non

Non sync. ..... 1

II. — Regina fol. 56.
Die iij Marcii 1465.
Captuni est in maiori Consilio : Quod Rector monouasie elegendus de tubus in tres
menses habere debeat salarium suum a loco nostro stalimnis et quum facile accidere p
per magnas impensas <[Ua8 idem stalimnis locus habet quod inde salarium ipsuni suum
habere non posset . . . Vadit pars quod in quantum idem rector noster monouasie a
Stalimnis insula salarium ipsum suum habere non posset juxta formam presentis electionis
sue a camera nostra crete illud percipere debeat sicuti conueniens et honestum est de
tribus in tres menses juxta formam presentis ipsius.

fDe parte 573

I >e non 39

N ' > 1 1 syin ere. . . . . 4-J

W. M.

U.S. Vol.. XXVII.

THE CENTRAL GROUPS OF THE PARTHENON PEDIMENTS.

There is probably no subject in Greek archaeology which has afforded
material for so much discussion as the identification and arrangement of the
sculptures of the Parthenon, and at first sight it may seem presumptuous at
this time of day to bring forward new views. But, on the other hand, it
must be remembered that discussion cannot be wholly fruitless, even if it does
no more than bring us closer into touch with the work of Pheidias ; and even
this well-worn path is not a mere wandering up and down, but sometimes
leads us farther. Jf it is vouchsafed to us by taking thought only to add
one small fragment to the Frieze, the labour is not in vain. And so I have
the less hesitation in returning once more to the question of the central
groups of the two Pediments of the Parthenon.

Discussion on these two subjects has entered (as I shall hope to shew) on
a new phase since the discovery was recently made that we must remove one
of the figures commonly assigned to the East Pediment and place it in the
West. It is now, I believe, generally admitted that the supposed Nike of the
East Pediment does not belong to that Pediment at all, but is the figure shewn
in Carrey's drawing as in the West Pediment, next behind the Poseidon.
Mr. Arthur Smith has recently found among Lord Elgin's papers at Broome
Hall two further letters from Lusieri and Dr. Hunt, which make it finally
clear that the torso in question was found beneath the West Pediment. A
summary of the evidence, which seems to carry conviction, is given in the new
edition of the Parthenon Guide, p. 32.

That being so, it is fairly obvious that this torso cannot be identified
any longer as Nike ; for what should Victory be doing in the train of the
opponent of the victorious Athena ? Studniczka (Jahrb. 1904, p. 10) suggests
that the figure is Iris, communicating to the disputants the will of Zeus. But
the same objection applies equally here : in a composition where, as we know,
every detail had its value both as composition and also in relation to the subject
represented, what significance could there be in giving Iris such a place ?
The hurried movement of the figure suggests that she is advancing against
Athena; in any case both her position and her action would be appropriate
bo Eris, the personification of strife ; and if so, the expression used by Pau-
sanias in describing this Pediment acquires a new force when he speaks of
the ' Eris of Poseidon against Athena.'

THE PEMMENTS OF THE PARTHENON. 243

This, however, is beside the present purpose : the fact remains that this
very reasonable innovation leaves us without a figure of Victory in either
Pediment.

And yet, when we consider the requirements of the case, and the
Athenian habit of mind in relation to these subjects, as we can judge of it
from contemporary monuments, we must see that at this period, in the
representation of any of the strenuous moments of her being, the presence
of Nike is essential. On the Frieze, in the peaceful sociability of Athena's
surroundings, her helmet and spear are cast aside; but wherever her panoply
is donned, Nike cannot be far away. Even in the other, the pacific render-
ing of the dispute between Athene and Poseidon, as shewn, for instance, in the
Smyrna relief (Ath. Mitth. 1882, PI. I., Fig. 2), it is Nike who counts the
votes : and in an Athenian monument of the importance of the Parthenon it
is hardly probable that so essential a feature should be omitted.

A curious illustration of this tendency is shewn in a red-figure vase in
the British Museum (E 410;. It dates probably from the latter part of the
first half of the fifth century, and is therefore before the date of the Parthenon.
Here the birth is represented in the old traditional way, with the miniature
Athena leaping from Zeus's head : the artist has felt so strongly the necessity
of introducing Nike somehow that he draws her on the extreme margin of
the scene., and in defiance of the unities represents her as nearly twice the
size of Athena herself.

The necessity of Nike's presence in the West Pediment was felt long-
ago by the early commentators; thus Visconti identified the figure who
drives the chariot of Athena as a 'wingless Victory,' but that is a type
which is unfamiliar to the artists of the Pheidian period.

There is yet another reason why Nike must be predicated for the
Pediments. In the pedimental compositions of the fifth century, so far as
we have them, tradition demanded (as, for instance, at Olympia and Aeginai
that the culminating point of interest should concentrate in a single figure
in the centre, occupying almost the entire height from floor to apex. Sauer's
examination of the marks of attachment in the floors of the Pediment shews
conclusively that Pheidias adopted a different principle : in each Pediment we
have two important figures balancing each other on each side of the apex.
In the East Pediment Zeus and Athena balance as they do in the Frieze ; in
the West, Athena and Poseidon. But this arrangement leaves a gap in the
apex which must be filled.

What is more important still, we are left in doubt as to which is the
predominant figure of the composition. In the temple of Athena it would
surely be made clear that it is the goddess herself, and not Poseidon or even
Zeus, who holds the pride of place.

Both these difficulties are overcome by the introduction of a small
figure of Nike — not so large as to make her structurally difficult to insert,
"i to give her undue importance, but such as would shew clearly, by her
inclination towards Athena, that here is the important moment of time and
place. In both Pediments the rhythm and How of the composition takes

B 2

244

CECIL SMITH

the eye inevitably from the angles to the apex ; and here in the crown and
summit of things, though the balance till now, as it were, was even, the little
satellite of Athena, who herself is Athena Nike, comes with no uncertain
voice to decide the issue.

Fig. l. — Scheme of Restoration of Central Group of the E. Pediment (after

FURTWANGLER-REtCHHOLD, Pi,. 20), WITH SaUER's PLAN OF THE

Floor below.

In the East Pediment there is a certain appropriateness in the introduc-
tion of a miniature figure at this point: as we see from the vase pictures,
tradition before Pheidias's time demanded a seated Zeus with a miniature
Athena springing from his head. Whether Pheidias was or was not the
inventor of the new scheme adopted in the East Pediment, it was, we know,

THE PEDIMENTS OF THE PARTHENON,

245

an innovation; and it is possible that Pheidias may have felt that the small
Nike here would be, as it were, a concession to artistic tradition.

In the West Pediment, again, a Nike introduced to mark the issue
obviates a difficulty which has always been felt. Petersen, for instance

FlG. 2.

-Scheme of Restoration of Central Group of me W. Pediment:

AFTER THE KERTCH Vase (EXTANT FRAGMENTS INSERTED);

Saueu's Plan of the Floor below.

(Hermes, 1882, p. 131), expressed the belief that Pheidias had succeeded in
rendering victory and defeat in the action of the two deities without recourse
to the presence of judges, which he regards as unfitted for sculpture, although
the later literature assigned as judges the twelve gods or the Athenian people ;
but the attitude oi the groups on either side (as A. S. Murray pointed out) is

246 CECIL SMITH

certainly not judicial. One has only to look at Schwerzek's restoration of the
Pediment to see how meaningless the whole composition becomes when
explained in this way and with no central Nike.

Let us now turn to the monuments, and see how far the introduction
of Nike is borne out by them. But first I should like to remark in passing
that there is one class of remains which has, I think, been somewhat neglected
in this connection. I mean vase paintings. The Kertch vase (which I shall
come to presently) is mostly regarded as telling us little of the Pheidian
West Pediment, although it remains the only authority we have for it. There
is one class of Attic vases in particular, which date from just after Pheidias,
and which, oddly enough, are nearly all found in the Crimea, which are full
of suggestions of motives directly or indirectly borrowed from the Parthenon.
I see no reason why the humbler handicraftsmen who painted the vases should
not have been proud to reproduce the sculptural types which were their
national glory. I believe we shall find that this track is worth following and
I have here one or two examples which at least seem to throw light on the
central group of the East Pediment.

First, however, in order to recall the composition that we may
expect, let me call attention to the marble well-head from Madrid, which
is now generally accepted as giving the most satisfactory rendering of
the main features of Pheidias's composition. The positions of Zeus and
Athena are shewn by the marks still existing in the floor of the Pediment
to be approximately identical, but the question of comparative scale raises
a difficulty. The head of Zeus, though he is seated, is on the same level as
that of Athena, so that if he stood up he would be on a much larger scale.
But the artist of the puteal is only following out the principle of isocephalism
commonly observed in all frieze composition, and especially so in the Frieze
of the Parthenon. It does not at all follow — indeed, it is extremely unlikely,
that the same principle would be observed in a pedimental composition in
the round, for the carefully adjusted balance of right and left, a balance
minutely calculated in every other group of the Pediment, would thereby be
ujiset.

The true arrangement (Fig. 1) is shewn on a vase picture (Furtwangler-
Reichhold Taf. 20) in which, though the subject is not the Birth of Athena,
the central group is clearly a reproduction of the Pheidian. The types of
Zeus and of Athena are both precisely what we should expect of Pheidias :
note especially the helmet of Athena and the gorgeous woven peplos that
she wears. And one sees at once how admirably the whole is adapted to the
rei|uirements of the centre of a pediment. As, however, the vase artist is
not confined, as the sculptor was, to an angle on the upper border, he has
probably slightly modified his Nike, who would naturally (as on the puteal)
have flown more directly towards Athena. The figure of Athena, on the
other hand, which in the puteal moves rapidly to the right, is here in a more
probable position.

A curious detail of perspective, by the way, is worth noting: the vase
artist for no apparent reason has drawn the underside of Zeus's throne in

THE PEDIMENTS OF THE PARTHENON.

247

such a way as it would be seen from below. Is it possible that this is due to the
fact that the artist was actually so copying the pedimental group from below ?

A slight modification of the same group appears on another contempor-
ary vase (Fig. 3 1 ), where Nike is shewn 'in what is more likel y t o have been
1 From Comptes Bendus de la Comm. Imp. Archiol. I860, PI. II.

248 THE PEDIMENTS OF THE PARTHENON.

her true direction. An interesting feature of this vase is the introduction
of other Pheidian motives, which though modified are unmistakeable. On
the right we have a woman riding on a horse, which naturally suggests the
Xyx or Selene of the right-hand angle of the Pediment ; and on the left a
group of three women, who in their relative positions strongly suggest the
group of the three Fates.

Turning now to the West Pediment (Fig. 2), the most complete document
w r e have is the drawing made in 1683 for the Marquis de Nointel. The
drawing is badly executed and the figures overcrowded, hut it gives a good
general idea of the central group. Comparing this with the vase from Kertch,
we see how closely the artist has followed his original : the miniature temple
in the corner (a most unusual feature in a vase painting) can, I think, be
nothing else than the artist's shorthand method of acknowledging the source
of his design. It wall be noted that Athena turns partly away from her
adversary, and both are very definitely striking downwards : probably
Pheidias would have wished to make it clear that the hisrh o-ods are not
engaged in turning their weapons against each other. The existence of the
olive tree is attested by fragments that remain ; and like a bird out of its
branches comes the little Nike with the victor's wreath or riband for Athena.
Whatever the relative positions of the two Nikai in East and West may have
been, it is clear that there was scope for the avoidance of sameness in a com-
parison, just as in the general composition there is general responsion with
contrast of detail : so the one Nike flies free from the apex with partial
inclination to the right; the other flies with inclination to the left out of the
branches of an olive tree. The group of Athena and Nike reappear on
another vase of the same series.

The question that now remains is a practical one : Have we any frag-
ments which can be identified with these two Nikai, and how were the figures
attached ?

As regards the second point, the question has been considered by Dorp-
feld, a practical authority, who sees no inherent difficulty as regards the East
Pediment in supporting a Nike on metal standards which would be cramped
to the tympanum or floor; in the West Pediment the olive tree would give
plenty of scope for the concealment of such cramps. If the figures of Nike
were of marble, it is possible that among the fragments of large and small
wings in the Acropolis Museum which are said to belong to the Pediment
sculptures portions of their wings may yet be found. I am inclined to think
that the Nikai may have been of some other material, such as bronze — among
the many bronze enrichments of the sculptures throughout the Parthenon
such a material for a small figure would not be out of place ; and in the
vase pictures of the time the practice commonly occurs of detaching this
class of figure (Eros or Nike), when in a central position, by a wash of white
or of gold.

( !ecil Smith.

J. H S. VOL. XXVII. (1907). PL. XVII

J. H. S. VOL. XXVII. (1907). PL. XVIII.

J. H. S. VOL. XXVII. (1907) PL. XIX.

J H S. VOL. XXVII. (1907). PL. XX.

THROWING THE JAVELIN.

[Plates XVII-XX.]

A. — The Javelin and tfa Amentum^

The javelin used in Greek sports is called variously a/ccov, ukovtlov,
fieac'r/Kv\or. ai~/vvvos, uttoto/xc'k;. The latter term, defined by Hesychius
as o-xt&v kcii ukovtlov irevrd9\ov, appears to denote merely a lath or stick,
and -apparently describes the javelin as represented on the vases. It is
merely a straight pole, in length nearly equal to the heigh; of a man. though
occasionally lunger, and about the thickness of a finger. It is one of the
commonest objects in palaestra scenes, whether in use, or carried in the hand,
or planted in the ground singly or in pairs apparently to mark the line from
which the athlete is to jump or throw the diskos. 1 These rods were
formerly described as jumping poles: but of the pole jump there is no
evidence, and the fact that they are precisely similar to javelins which are
actually being thrown, and that they often have the throwing-strap or amentum
attached, proves that they are nothing more than javelins. At the same
time there is no reason why they should not have served a- measuring-rods
or Kavove? for measuring the jump, a use which is perhaps represented on
the British Museum kelebe B. SGI. published in vol. xxiv. of this journal,
p. 18o.

The athletic javelin is in the vast majority of cases pointless. < >u early
black-figured vases, such as the kelebe just mentioned, or the B. M. lekythos
B. 576 published in this volume, PI. II, the javelin is represented by a black
line which does seem to taper at the ends, but this is a mere accident of
technique, the natural result of a line drawn rapidly with a single stroke of
brush or pen. On the red-figured vases and Panathenaic vases, the rod is of
uniform length and usually ends square. Sometimes indeed it appeal
have a blunt cap or ferule, indicated by a thickening of the end. or by a black
patch, or bylines which represent the binding by which it is attached.-' Such
are the javelins and spears which Xenophon recommends cavalry soldiers to
use- in practice, provided with a round end ia^aipw/xeia) '• which corresponds
to the button on the modern foil or bayonet. Such a cap served not only for
protection but to give the necessary weight to the head of the javelin,

1 v. p. 11 ol this volume: also vol. xxiv. Fig.

p. 186. 3 !'■ viii. 10.

2 Juthner, Antih Titrngcrathc, p.

250

E. NORMAN GARDINER

without which it would not fly properly. It was probably universal, the
omission of the lines or shading that indicate it being due to the vase-
painter's carelessness. We cannot however ascribe to the same cause the
omission of the point, seeing that in hunting and fighting scenes the javelin
has almost invariably a long leaf-shaped metal head ; and we may therefore
safely conclude that the blunt javelin was generally used for practice, espe-
cially for distance throws. For target practice sharp javelins were naturally
used, as is proved by the speech of Antiphon in defence of a youth who, missing
the mark, accidentally hit and killed a companion. 4 Moreover, on two of the
three vases which represent javelin throwing on horseback at a target, the jave-
lins have all long leaf-like points, such as we see in hunting scenes. 5 A similar
head is roughly represented on an early b.-f. hydria in the British Museum
B. 326, where we see two athletes carrying javelins; but of the five weapons only

Fig. 1.— Various Methods of attaching the Amentum.

a. Vase of Hieron (Pollak, Zicci Vasen Hierons). b. B.M. Kylix, E 58. e. B.M. Lekythos, E 698.

d. B.M. Kylix, E 96. c. Alexander Mosaic : Jiithner, Fig. 38. /. B.M. Amphora, E 316.

one is pointed. Somewhat similar is the javelin on a well known Chiusi wall-
painting, while on the Lateran mosaic occurs a barbed javelin head. 1 ' On the
Berlin bronze diskos the javelin has a long thin point attached to it by
a socket, and similar points occur on a few r.-f. vases. 7 These however are
but isolated examples, and the enormous preponderance of the blunt javelins
justifies the conclusion that though for target practice the sharp javelin was
used, the blunt one was preferred for distance throwing, and that down to the
close of the 5th century distance throwing was more general than throwing
at a target.

Whether pointed or blunt the javelin was evidently a light object, and
Anacharsis contemptuously contrasts it with the more serviceable weapons

4 Tctralogia, ii. 2.

5 v. infr. Fig. 16 and PI. XX.

6 Jiithner, op. cit. p. 39, Fig. 33 ; Schrciber-

Anderson, Atlas, xxii. 9.
7 Jiithner, Figs. 20, 40.

THROW] XG THE JAVELIN. 2 5 1

which are not carried about by the wind. 8 It was thrown by means of a
tlmiig called dyfcvXr), or amentum, fastened near the centre of gravity of the
javelin, which was therefore called /xeo-dy/cvXov. The amentum was a leathern
thong a foot or eighteen inches in length, if we mayjudge from the numerous
representations of an akontistes holding an amentum loose in one hand and a
javelin in the other.'-' It was bound firmly round the shafl of the javelin in
such a way as to Leave free a Loop 3 or 4 inches long, in wliich the thrower
inserted his first, or firsl and middle fingers. The point of attachment was
the centre of gravity, in the light-headed javeliu of the athlete almost in the
centre of the shaft, in the more formidable weapon of war or the chase
generally nearer the head. Its place varied also according as the javelin
was to be thrown for distance or at a mark. By putting the ameutum
behind the centre of gravity it is possible to increase the distance thrown,
but at a sacrifice of accuracy. Hence the amentum was detachable,
and the athlete fastened it to suit his taste shortly before use < m
the r.-f. hydria in the British Museum, K 164, published on p. 32 of
this volume, we see a youth sitting on the ground in the act of attaching
the amentum. On a Wiirzburg r.-f. kylix published by Jiithner
youth is bending down winding the amentum round the shaft, while he
holds the other end tight with his foot. It was as we shall see essential
that the thong should be securely fastened. The vase-paintings are
too minute to show precisely how the amentum was fastened, but tbey
suggest a considerable variety of ways, as will be seen from the examples
given in Fig. 1. The clearest example is that shown on the Alexander
mosaic in Naples. In every case it is only the actual loop which is Left
free.

The amentum served various purposes. In the first place it enabled the
thrower to give a rotatory motion to the javelin, wliich not only helped it to
keep its direction, acting on it in the same way as the rifling of a gun on a
bullet, but also increased its carry and its penetrating power. For this reason
the modern savage habitually puts a spin on his weapon by a movemenl of
the fingers, which however are not so effective as the thong. The manner in
which the amentum acts is shown in the illustrations reproduced from
Jiithner in Fig. 4 a, b. The carry was further increased by the additional
Leverage given to the thrower's arm. The amentum also served to mark
the point at which the javelin was to be grasped, a matter . . t considerable
importance in war or in the chase, when there is no time to adjust the
weapon carefully. In javelins and spears used by savages at the present
day this point is often marked by some sort of binding. Such binding
cle.uly shown on the long spear held by Athena on the British Museum r.-f.
amphora E 316, a portion of which is shown in Fig. 1 f. Lastly, the amentum

s Lucian, Anackarsis, 32. ivhere and commonly described as com

:l Juthuer, p. 40, Fi<, r s. 34, :!.">, 36. Jiithner are merely amenta, somewhat misdrawn.

proves conclusively that the objects represented Op. cit. Fig. : '>7.

on the Panaetius kylix ('-. infr. Fig. 12) ami

252

E. NORMAN GARDINER

afforded a convenient handle, also a matter of some practical importance.
Hence it was sometimes attached to long spears not intended primarily for
throwing. Spears are sometimes provided with a short loop or sling for
this purpose only. 11

From these considerations it is obvious that the amentum was not as is
sometimes stated the invention of the gymnasium, but was adopted by
the gymnasium from war and the chase. This is abundantly proved by the
monuments. Whether it was used in Homeric times, is uncertain. The prin-
ciple of the sling was certainly known to the Homeric shepherd, 1 ' 2 and besides
the SoXixoaKiov ey^o? of the chieftain, there was apparently a lighter and
shorter weapon, the alyavei], which is closely associated with the bow, and
like the bow was used for hunting, and by the common soldiery in war and in

Fig. 2.— Interior of B.-F. Kylix. B.M. B 380.

sport. 13 But there is no evidence in Homer that the principle of the sling
was applied to the alyavet]. The warrior vase from Mycenae 14 however
clearly shows two types of spear, a long spear carried with clenched fist, and a
short spear raised almost at arm's length behind the head, with the point
somewhat downwards, in a position commonly represented in hunting scenes.
The hand is sharply pointed as if the fingers were extended, and a com-
parison of the way in which the spear is held with the hold shown in
Figs. G and 7 confirms Jiithner's view that the artist intended to represent
a weapon thrown with the amentum.

From the sixth century onwards the amentum was used in war, in the

" Possibly the Hasta Ansata is such, but
v. infr. p. 255. Such spears are known among

tin- Indians, r. Egerton Arms of the Indians,
Figs. 2, No. 6 ; 72, No. 2."..

12 r. p. 5 of this volume (the- Ka\avpo\p) ; ep.
//. xiii. 600. Slingers are represented on a

fragment of a silver vessel from Mycenae.

13 77. xvi. 589, ii. 774. Wc may note that
both Odysseus and Nestor speak of their skill
Sovpi, not aKovricj). 00. viii. 229 ; //. xxiii. 637.

14 Schliemaun-Schuchardt (Eng. Trans.),
Figs. 284. 5.

THROWING THE JAVELIN. 253

cliase, and in athletics. A few examples will suffice. On a Corinthian b.-f.
lekythos in Berlin showing the various pieces of a hoplite's armour there stands
side by side with the long spear a shorter weapon provided with a sort
of loop. 15 The use of the amentum is clearly shown on a Chalcidian
b.-f. kylix in the British Museum (Fig. 2). As is often the case the artist
has made a mistake in drawing the fingers : the first and middle fingers
should have been passed through the loop, not the other two. The warrior
is perhaps about to throw the spear with a short underhand throw, a
throw in which certain savages are extraordinarily skilful and for which
the amentum woidd be most useful. Another Corinthian vase, in the
British Museum, B 37, shows a delightful hunting scene. Several
javelins fitted with amenta are seen sticking in the boar's back, clearly
proving that the amentum was fixed to the javelin and did not remain
in the thrower's hand. On yet another archaic vase, to which Mr. Cecil
Smith has called my attention, a javelin fitted with the amentum is flying

Fig. 3. — From the Feanc/ois Vase. (From Furtwangler, Vasenmalerei XIII.)

through the air, and the artist to produce a sense of velocity has given
the shaft a wavy appearance. 10 Lastly on the Francois vase we see a
pair of warriors with their fingers in the amentum about to throw their
spears (Fig. 3), who in the position of the hands and fingers, and the whole
attitude, closely resemble the akontistes in Fig. 10, save that on the latter vase
the head is turned backward, a position obviously inadvisable for a hunter
or warrior. As an early example of the amentum in athletic scenes we
may take the akontistes on the b.-f. stamnos in the Museo Qregoriano
(Fig. 7).

A javelin thrown by a thong is necessarily a light weapon, but though
light the akontion used in war and in the chase was decidedly formidable,
and could be used effectively not only for throwing but for stabbing.
For the latter purpose it could be held either with clenched fist or with
the amentum. Some of the figures on the Francois vase for example (Fig. 3)
have their javelins raised behind their heads like the warriors on the Myc

'•' Jiithner, Fig. 51, Berlin 3148. " ; Arch. /. 38 .'1. X.

254 E. NORMAN GARDINER

vase as if to stab ; the hands are clenched and there is no sign of the
amentum. The type is a common one on vases of all periods. That the
amentum would afford a useful grip both for throwing and stabbing is clear
from the hunting scenes represented in Gerhard's Apulische Vasenbilder,
PL A. •

If the akontion could be used for thrusting, the long spear could
on occasion be thrown. The Homeric warrior sometimes hurled his
hoXixocncLov e'7%09, and so did the hoplite of the fifth century. The long
spear is even represented sometimes with an amentum, for example on
the British Museum r.-f. lekythos E 698, on which is drawn a female
figure, possibly Eudaimonia, holding in her hand a long spear, or on the vase
of Hieron from which the detail of Fig. 1 a is taken. 17 How far this practice
was general, it is impossible to say. The amentum is a detail only occasionally
inserted by the vase-painters whether on javelin or spear, and even when
inserted the thin lines which indicate it are very liable to wear away.
Certainly the primary use of the long spear was for thrusting : the hoplite
could not afford to risk its loss and would rarely throw it. The amentum
therefore if generally attached served probably rather as a handle than as
an amentum proper.

The lono- heavy spear was the weapon of the fifth century : the real
importance of the javelin dates from the closing years of the Peloponnesian
war, when the value of light-armed troops and cavalry began to be realized.
The light-armed troops were mostly mercenaries, Lydians, Mysians,
Arcadians, Aetolians, Thessalians, Thracians. 18 All these races were skilled
in the use of the javelin. That the peltasts threw the javelin by means of
the amentum is clear from Xenophon. In the passage of the Ten Thousand
through the mountainous territory of the Carduchi, the Greeks he tells us
picked up the long arrows of the enemy and used them as javelins fitting-
thongs to them (evay/cvXcovres). 10 Elsewhere he orders the peltasts guarding
the rear to advance with their fingers in the amentum (8niyKv~\.i<r/jt,evou<i)
ready to throw.

The javelin was not confined to the mercenaries : at Athens it was the
special weapon of the ephebos, who is generally represented holding a pair of
javelins. PL XVII. gives a typical picture of the Athenian ephebos.
In the third century special trainers called uKovnarat were engaged
to train the epheboi in the use of the javelin. 20 Competitions were
held at Athens and elsewhere in throwing the javelin both at a mark
and for distance, both on foot and on horseback.- 1 In the vases representing
the latter contest the amentum is not shown, but its use is implied by

J < The anient 11111 occurs frequently on 19 Anab. iv. 2. 28, iv. 3. 28, v. 2. 12.

Hieron's vases, e.g. on a r.-f. kotyle represent- '-'" Ditt. Syll. 2 ii. 520, 521 ; cp. 522 of Ceos,

ing Achilles and Briseis in the Louvre, M.d.I. 523 of Teos.

vi. 19; on a kylix representing Theseus and -' Ditt. Syll- ii. 522, 668, 670, 671, 672,

Aethra, Hermitage, 830, Harrison and MacColl 673,671; Inscr. Or. Gr. 339; I.G. vi. 441;

Greek I tings, PL XXII. Ath. Mitth. xxx. 1905, p. 213.

13 Dar.-Sagl. s.v. jaculum, p. 594.

THROW INc THE JAVELIN. 255

the fingering of the javelin, and this conclusion is confirmed by the representa-
tion of a cavalry skirmish on an embossed swordbelt found at Watsch
in Carinthia, where the spears flying through the air are all provided
with a loop, 22 or that on a vase from the Acropolis described below (Fig. E

B. — The Distribution of the Amentum.

The Romans are said to have borrowed the amentum from the Greeks. 23
But the evidence of its distribution seems to render this statement highly
improbable. The amentum was widely known in Italy at an early date. It
was certainly known to the Etruscans, being represented on an Etruscan
warrior's spear in a tomb at Caere, while in another tomb at Ohiusi an
Etruscan athlete is depicted in a typical position, putting his fingers through
the loop.- 4 From other tombs and from vases representing Italian warr -
we learn that the Samnites and Messapians used the amentum.- 5 Avery
interesting painting found in a tomb at Paestum represents a fight between
two warriors, each armed with a shield and two javelins, fitted with a sort of
semicircular loop. 2 " Two of the javelins have been thrown, one of them is
sticking in the left hand warrior's shield. The other has pierced right
through his opponent's calf, so that its point projects on the other side. With
their remaining javelins they are preparing to stab one another. They hold
the javelins with clenched fists, but in only one case is the thong visible,
forming a loop over the holder's hand but certainly not used as an amentum
proper. Hence the javelins have been identified with the hastae ansatae
mentioned by Ennius. But whether the object represented is an amentum
or ansa, or whether the amentum is the same as the ansa or different, we
cannot say. I am inclined to think that the amentum did at times sen
a handle or ansa, and that the object represented is intended for the
amentum, for which ansa is merely another name. But though there is
evidence of the wide distribution of the amentum in Italy at an early date,
there is no evidence of its use in the Roman armies. It is stated indeed that
the hasta velitaris used by the light-armed troops was thrown by the
amentum, but I do not know on what evidence. Certainly the characteristic
Roman weapon was the heavy pilum, and Livy in his account of the battle
of Magnesia expressly contrasts tin' heavy weapons of the Roman soldiers
and the light weapons of Antiochus' forces. 27 The damp, he says, had no
effect on the heavy pila and swords of the Romans, but it had softened the
bow-strings, slings, and the 'amenta jaculorum' used by the king's soldiers.
The light javelin was. as we have seen in Greece, essentially the weapon of the
hunter^ and the light-armed soldier, and the strength oi Rome lay in her

/, . ireh. 1884, PI. III.: Juthner, M Dar.-Sagl. s.v. amentum, Figs

tjl 62.
28 F. Krause, S^ileudervorrichMmgen fur

IVurfwaffen in I "' xxxvu - 41 -

1902, pi'. 121 sq.

256 E. NORMAN GARDINER

heavily armed legionaries, her light-armed troops being supplied mostly by
allies and at a later period by mercenaries. Among the enemies of Rome
Ave undoubtedly find a weapon thrown by means of the amentum. This was
the tragula used by the Spaniards at the time of the Second Punic War. In
Caesar's time the tragula is the weapon of the Gallic cavalry, serving as
mercenaries in his army. 28 From this time we have abundant evidence of
the use of the amentum in the Roman army. We find traces of it in the
Roman weapons discovered at Alise Sainte Reine, and we even find it
represented on the Roman legionary's spear." 20 But it is unnecessary to go
into details of this period. Enough has been said to show that the amentum
was widely distiibuted over Italy, Gaul, and Spain, and that Rome probably
adopted it, not from Greece, but from her own allies and subjects. To suppose
that its use spread from Greece to Rome, and thence to Gaul and Spain, is
surely inconceivable.

This view is supported by evidence from other parts of Europe. We
have already found the amentum at Watsch in Austria. A comparison of the
light javelin heads found in large numbers at La Tene with those found at
Alise renders it likely that the weapons to which they belonged were thrown
with a thong. 30 From Vergil we gather that the use of the throwing thong-
was a Teutonic custom." 1 In his catalogue of the Latin forces he mentions
two apparently similar weapons, the aclys and cateia. The former was
thrown by the ' lento flagello,' the latter ' ritu Teutonico.' Unfortunately we
do not knew the exact nature of the weapons mentioned. Proceeding
further north-west we find conclusive proof of the existence of the amentum
in the early iron age of Denmark. 3 ' 2 Remains of it have actually been found
at Nydam. The spears found are from 8 to 10 feet long. On the middle of
the shaft are often visible a number of small bronze rivets, between which
the cord was fastened, or a bronze mounting to mark the point of balance.
In some cases the cord was found still fastened between the rivets. Lastly we
find the amentum in Ireland. 33 In ancient Irish story it is frequently mentioned.
Thus in the battle of Moyreth ' Cuanna pressing his foot on the solid earth
put his finger in the string of his broad-headed spear and made a cast at
Congal.' This loop — called suanem, or suaineamh — was made of silk or
Max, and the laigan or spear to which it was attached is said to have been
brought into Ireland by Gaulish mercenaries in the fourth century B.C.
A most interesting survival of this old Irish spear with its loop is seen
on a picture of Captain Thomas Lee painted in 1594, now in the possession
of Lord Dillon, a print of which is in the Ashmolean Museum at Oxford.

The fixed amentum seems practically unknown to the modern savage.
The only example that I can find is a javelin from, the Pitt Rivers collection
no. 217, now in the Ashmolean Museum. It is described as coming from

Livy, xxi. 7. 10. xxiv. 42. 2; Caesar 3 - ('. Engelhardt Denmark in the Early Iron

B.C. i. 2*i, v. 35, 18 ; B.C. i. 57. Age, p. 56, Pis. X. 5, 0, XII. 2.

-> Dar.-Sagl. s.v. hasta, p. 39, Fig. 3729. 33 P. W. Joyce A Social History of Ireland,

Desor Les palafittes, p. 86. i. 113; O'Curry Manners and Customs of

31 Aen. vii. 730. 741. Ancient Ireland, I. ccccxliv.

THROWING THE JAVELIN. 257

central Africa, where it is probably a survival of the javelins used by the
Roman mercenaries occupying Africa. 34

Another type of throwing thong, the ounep, is used by the people of New
Caledonia and New Hebrides. It is a thickish cord about 6 or 8 inches long
with a loop at one end for the finger and a knot at the other. The spears
are 9 to 12 feet long with a slight projection just behind the centre of
gravity, behind which the cord is placed and twisted over the knot in such
a way as to untie as the spear leaves the hand, remaining itself in the
thrower's hand. An illustration of this is taken from a drawing displayed
in the Ethnographical Gallery of the British Museum (Fig. 4). A
combination of this thong with the throwing stick is used in New Zealand.
The throwing stick is by far the commonest means of increasing the

b d

Fig. 4. — Illustration* of the Use of the Throwing Thong.
a, b. Jiithner, Figs. 47, 48. c. B.M. B 134. d. The Ounep of New Caledonia.

throw of a spear. It is widely used in Australia, Melanesia, Central America,
and among the Eskimos, but is unknown in Europe, though a similar
implement of bone was apparently used by Palaeolithic man in France. 35

To sum up, the fixed amentum is an exclusively European invention. It
is found throughout Greece and Italy, in Spain and Gaul, in Central Europe, in.
Denmark, and in Ireland. The light javelin to which it belongs is essentially
the weapon of the less highly civilized peoples. It is a weapon of the chase, a
weapon of the common people, but it plays little part in the heavily equipped
citizen armies of Greece and Rome. Both in Greece and Rome it comes into
prominence with the organization of light-armed troops, and then it is chiefly
the weapon of the troops of subject states and mercenaries. Under these

34 Bcthnal (h ■■ l y 77. ionales Arehiv, l>\ B.M.

p. 40 ; Pitt Rivers 7 on of Culture, to the Stone Age, p. 49.

pp. 182-4.

H.S. — VOL. XXVII.

258 E. NORMAN GARDINER

circumstances it is hardly conceivable that its use should have originated in
Greece, and we are forced to the conclusion that the amentum was a device
of the peoples of Central Europe, and in the course of their wanderings was
carried by them throughout the Southern and Western portions of the
Continent.

Experiments have amply established the practical use of the amentum
with a light javelin. In the experiments made by General Reffye for the
Emperor Napoleon it was found that a javelin which could be thrown only
20 metres by hand could be thrown 80 metres with the amentum. Accord-
ing to Jtitlmer an inexperienced thrower increased his throw from 25 to 65
metres. These records are with a light javelin. The javelin used in the
recent Olympic games was of the Swedish type, without the amentum, and
weighed 800 grammes or 2 lbs. Lemming, the winner, threw it 53*90 metres,
but Colonel V. Balck informs me that in Sweden he has thrown nearly 55
metres. The old Greek javelin must have been a much lighter weapon.

C. — The Manner of Throwing the Javelin.

In spite of frequent carelessness and mistakes in the drawing of hands
and fingers, in spite of the frequent omission or disappearance of the amentum,
the vases leave no doubt as to the method of throwing the javelin with the
amentum. Two things are necessary : the amentum must be firmly attached
to the shaft, and the loop must be drawn tight by the fingers before the
javelin is thrown. We have already seen the akontistes in the act of fasten-
ing the amentum. On a r.-f. psykter (Fig. 5) we see a group of javelin
throwers preparing to practise under the supervision of a paidotribes and his
assistant, while two other paidotribai are obviously giving instruction in
wrestling. Two of them are testing the bindings: resting one end of the
iavelin on the ground and holding it firm with the left hand, they pass their
right hands along the shaft so as to see that the binding is secure. A third
holding the javelin in the same position is about to pass his fingers through
the loop. The loop has disappeared, but is clearly indicated by the position
of the hand. A fourth has already inserted his fingers through the thong,
and raising the javelin horizontally to a level with his breast, presses it for-
ward with his left hand so as to draw the thong tight. The attitude of this
figure has caused quite unnecessary difficulty : it is a perfectly natural posi-
tion, from which by a half turn to the left any of the preliminary positions
which we shall now describe may be reached.

Dr. Jiithner distinguishes two types of javelin throwing, one in which
the javelin is pointed more or less upwards, the other in which it is hori-
zontal. The distinction, in spite of de Kidder's denial, is a real one, though
Jiithner has not grasped the full meaning of it. The so-called horizontal
throw is the throw of war or the chase, the other the throw of athletic
competitions. In the latter, distance is the one and only object, and the
thrower may take his time : in the former distance is only a secondary con-

THROWING THE JAVELIN.

259

Bideration, compared with force and accuracy, and everything depends on
rapidity of action. It is the difference between throwing in a cricket-ball to

FlG . 5.-R.-F. Psykter OF PHINTIAS. (Anlilx Dcnkmdlcr, ii. 20.)

the wicket and throwing a cricket-ball in a competition. An examination of
the two types will make this clear.

(a) The Practical Style,

The soldier or the huntsman must have his javelin ready for use at
a moment's notice. He therefore carries it with his lingers passed through
the loop, ^W,eW. He may carry it horizontally by Ins Side as does

260

E. NORMAN GARDINER

the warrior in Fig. 2, a somewhat cramped position owing to the fact that
the arm has to be turned outwards. The freer and more natural position is
with the arm bent and the javelin sloped over the shoulder or across the body,
the point downwards. From this position he can draw it back as does the other
youth from the same vase, or raise the elbow so that the javelin is level with
the head, an excellent position for taking aim. This manner of holding the
javelin is implied or represented on many hunting or battle scenes. It is
equally serviceable on horseback and on foot. But the best examples of it
are on the two Panathenaic amphorae (PL XVIII and Fig. 6). On the British
Museum vase the athlete who leads the procession carries his javelin at the
slope, the other akontistes has raised it horizontally. On the Leyden amphora
the javelin is still sloping slightly downwards. This position, with the javelin
poised on a level with the head, which we may call ' the carry,' is the
natural preliminary position for starting, whether the thrower uses the

Fig. 6. — Panathenaic Amphora, Leyden. (Arch. Zcit. 1881, IX.)

amentum or not. In the latter case it enables the thrower to balance
his weapon properly. In a photograph which Mr. Bosatiquet sent me of
Lemming, the winner in this competition at the last Olympic Games, the
latter is standing with his javelin poised in this very position. The javelin
may be kept in this position during the run or may be at once drawn back.
On the two vases in question the thrower certainly appears to be running,
but the position of the left arm clearly proves that he is not in the act of
throwing. Where time was no object, as in sports, the thrower might before
starting to run adjust the javelin by pressing the point back with his left
hand so as to draw the amentum tight, and this movement is represented on
a b.-f. stamnos in the Museo Gregoriano (Fig. 7). From the 'carry' the
thrower immediately before the throw draws back his arm in the manner
represented on the Francois vase (Fig. 3). In the actual throw the move-

THROWING THE JAVELIN.

261

nient is reversed, arm and spear travelling back through the same positions
again, saving that when the amentum is used the ringers loose their hold of
the spear which (even before the hand reaches the level of the head) is hell
merely by means of the thong. This is clearly shown on a most interesting
b.-f. vase from the Acropolis, the upper zone of which represents a battle of
chariots and the lower zone a cavalry fight between archers and javelin
throwers. One of the latter is reproduced in Fig. 8 from a photograph
of the vase for which I am indebted to the kindness of Drs. Wolters
and Graf, who are engaged in the publication of these vases. 36 Perhaps
the moment before the actual throw is represented on a Panathenaic
amphora in the Mnseo Gregoriano. :; " The attitude of the javelin thrower
is very similar to that shown on the two Panathenaic amphorae
mentioned above. The action however is decidedly more vigorous, the
body is inclined slightly forwards, and the left hand instead of being

Fig. 7. — B.-F. Stamnos. (Museo
< rregoriano II. xvii. )

Fig 8.— B.-F. Vase. Acropolis, Athens

raised is swung backwards, while the spear points slightly upwards. The
only difficulty in this interpretation is that the spear still rests between
the finger and the thumb, and is not as it should be at this point held
only by the amentum. But careful though the early vase-painters are
in details, the realism of the Acropolis vase is certainly exceptional, nay
as far as I know unique, and I am inclined to think that the artist
of the Gregoriano amphora did intend to lepresent the actual moment of
the throw. The three amphorae and the other vases discussed all belong
to the sixth century. The style of throw represented is typical of the black-
figured vases and quite distinct from that which we shall find ueneral on the
red-figured vases of the fifth century. It is the practical style of the chase
and war adapted to the palaestra, and in the fifth century, when owing to the

A ropolia vases, 568, cp. Jiithner, p. 52. variation occurs ou an archaistic r.-f. amphora

lft«. Greg, xliii. 2. b. The other figures figured by de Witte PI. XXIV,

are a diskobolos and an official. A slight Lug close to the side.

262

E. NORMAN GARDINER

development of the heavy-armed hoplites the light javelin temporarily lost its
practical importance, this style was superseded by a purely athletic style.

Before proceeding to discuss the latter we must deal with a question
which naturally arises. If the style of throwing is that of the chase and
war, does it not follow that these vases represent throwing the javelin at a
mark rather than for distance ? The question is of importance in connexion
with the nature of the javelin competition in the pentathlon. At first
sight the general attitude seems in favour of throwing at a target ; but the
care which the artists take to emphasize the fact that the spear point is
blunt is conclusive for a distance throw. There is no evidence at any period
for any kind of target or mark for which a blunt spear could be used. More-
over, both in sport and war distance and force are no less important than
accuracy, and it is natural that as long as the javelin was regarded as

Fig. 9.— R.-F. Kylix, Munich. 562 A. (After Juthner. )

a military rather than as an athletic implement, it should be thrown even in
an athletic competition for distance in the style most practically useful. 38

(b) The Athletic Style.

The purely athletic character of the style depicted on the red-figured
vases is obvious from the most casual inspection. Till the actual moment of
the throw the head is turned backwards, the eyes fixed on the right hand, 39
a position absurd alike for war and for the chase, and for throwing at a mark.
After carefully adjusting and testing the amentum in the manner described
and passing his first finger or first and middle fingers through the loop the

38 I have received confirmation of this from
a friend who has long resided in Central Africa.
The natives in war and hunting throw spears
much in the style described above, though
without the amentum, and in throwing the

spear for distance in sport they use the same
style, never the purely athletic style described
below.

39 Cp. Figs. 9, 10, 11, 14. 15.

THROWING THE JAVELIN.

263

thrower extends the right arm backwards to its full length, while with his
left hand opposite his right breast he holds the end of the spear and pushes
it backwards so as to draw the thong tight. This is the moment depicted
on the B. M. amphora E 256 (PI. XIX). The javelin is sometimes held
horizontally, more commonly sloping with the butt-end almost on the ground. 1
As the thrower starts to run he draws his right arm still further backward>
turning his body sideways and extends his left arm to the front. On tin-
Munich kylix (Fig. 9) the youth on the left still holds the point of the javelin
in his left hand, the youth on the right has just released it. On the Berlin
kylix (Fig. 10) the left arm is fully extended. From the position of the head

Fig. 10.— R.-F. Kylix, Beblin. 3130. (After Hartwig, Meisterschahn, PL XLVI.)

and body it is obvious that the violent, rapid run of which some authors speak
is an impossibility. Just as in throwing the cricket-bali the run consists of
a few short springy steps. Immediately before the throw a farther turn of
the body to the right takes place, the right knee being well bent and the
rio-ht shoulder dropped, while the hand is turned outwards so that the shaft rests
on the palm of the hand. This attitude is vividly depicted on a Torlonia
kylix, the illustration of which is taken from Jtithner (Fig. II). 41

* It need hardly be said that there is no
evidence for the ' tir en haut' of which de
Riddt-r speaks. Dar.-Sagl. s.r. jaculum. In
the r.-f. kylix from the Louvre (Dar.-Sagl.
Fig. 252, Schreiber, Atlas sxii. 8) which he
cites the angle of the spear hardly differs from
that in Figs. 0, 11, 14. Generally speaking the

higher the throw, the greater the carry. Cf.
Xenophon de '■ xii. 13, quoted below

p. 271.

41 Cp. the directions for spear-throwing and
the illustrations given by Col. V. Balck in his
Lekar och Idrottsojningar, p. 126.

264

E. NORMAN GARDINER

A variation of style is represented in Fig. 10. Here the javelin instead
of pointing upwards is almost horizontal. A similar variation has been noted

in the preliminary position. The

,&'

attitude closely resembles that adopted
in throwing at a target, and we might
be tempted so to interpret it. But
this interpretation is put out of court
first by the position of the head,
which has been already noticed,
secondly by the fact that in all the
vase-paintings of this type the javelin
is blunt. This type occurs on the
British Museum bronze diskos, which
is figured by Jiithner p. 28. The
javelin is represented by a single line,
which is misinterpreted in the cata-
logue as a cord. The drawing of the
akontistes is however so coarse and contrasts so strongly with that of the
diskobolos on the other side, that Mr. Cecil Smith has come to the con-
clusion that it is spurious. If genuine, the line can only be part of an

Fig. 11.— R.-F. Kylix, Torlonia 270 (148).
(After Jiithner.)

Fig. 12.— R.-F Kylix, Munich 795. (From Arch. Zcit.)

unfinished drawing of a javelin. The closest parallel to the diskos is the
vase represented in Fig. 10.

TllliOWINd THE JAVELIN.

265

The actual throw is very rarely shown and the artists who attempt it are
hopelessly confused. For example, the central youth on the -Munich kylix
(Fig. 9) is clearly intended to be throwing the javelin to the right, but the
lingering of the right hand is only compatible with a throw to the left. Not
much better is the drawing on the Panaitios kylix (Fig. 12). The general
attitude is good and lifelike, but the position of the hand is hopeless and the
amentum is conspicuous by its absence. The carelessness of the red-figured
vase-painters as to the amentum is in marked contrast with the carefulness of
their black-figured predecessors. For the former the typical positions of the
akontistes are the preliminary positions described above, which are repeated
with little variation till they become merely conventional. Moreover, whe]
in the black-figured vases the amentum is inserted in black in the same way

Fig. 13. — R.-F. Amphora, .Munich. (From Furtwangler's •' XLV.

as the spear itself, in the red-figured vases it has to be added in some other
colour, usually white or purple, after the drawing is finished. Hence this
detail tends to be omitted altogether, and if inserted is the first to become
obliterated.

Occasionally we find a type that reminds us of the black-figured
On a r.-f. 'kylix' in the Museo Gregoriano reproduced in Klein's E
we see a youth striding vigorously forwards with his javelin raised in his
right hand level with his head, and his left hand swung backwards. The
same type occurs on a kylix of Epictetos in Berlin. 42 The - aergy of the
action, which on the latter vase is encouraged by the strains of a flute, seems
i" suggest that the actual moment of the throw is represented. Bui

'-' ifus. Greg. Lxix. Klein, p. 291 ; Berl. Vas. 2262, Gerh. ././". 272.

2GG

E. NORMAN GARDINER

that the right leg is advanced seems decisive against this view, unless we can
suppose that the javelin is being thrown not for distance but at a target, of
which, as I have said, there is no evidence.

Similar but less vigorous is the drawing on the Munich amphora
(Fig. 13). Here again the right hand, which in the other two vases is
hidden by the head, is quite impossible, the wrist being curved over the
shaft instead of being bent back underneath the shaft. It seems safer then
to regard this type as representing the run.

A few small points remain. Was the javelin ever thrown without a run ?
That it was usually thrown with a run is obvious from the vase paintings, but

a drawing published by Jiithner from the
Apparat des ram. Instituts (Fig. 14) proves
that the standing throw was sometimes
practised, the attitude being evidently
borrowed from that of the diskobolos.
Possibly the Torlonia kylix (Fig. 11) may
also represent a standing throw.

Secondly, was the javelin thrown with
the left hand as well as with the right ? Plato
recommends the training of both hands
alike, and the fact that the Greek carried
two spears, often one in either hand, ren-
ders the suggestion probable. But the
only direct proof of a left-handed throw
is on a kylix of Nicosthenes, in Berlin. 43
Even if a left-handed throw was practised in the gymnasia, there is no
evidence of it in competitions.

Lastly, was the javelin ever thrown without the amentum? The only
evidence is derived from the position of the hand and the omission of the
amentum on the vases, and this evidence is, as has been explained, too
untrustworthy to warrant us in asserting that it was so. Here again common-
sense tells us that the Greek athlete, used as he was to the amentum, would
not have rejected its help in competitions.

-II -F. Kylix, Rome.
Jiithner.)

(After

D. — Competitions in Javelin Throwing.

In the games of Patroclus javelin throwing is a separate event. Here
and in all other passages where it is mentioned in Homer as a sport the
competition is for distance only. 44 Throwing at a mark may possibly be
implied in the association of javelin throwing and archery, 45 a combination
which meets us again in fourth-century inscriptions, and Pindar definitely
describes such a competition in the legendary Olympic games celebrated by
Heracles,

CIKOVTL fypuCTTOyp i'fkdOe (TKOTTOV. 01. X. i\.

4:t Kranse, xviii. //, ] 1 r. ; Berl. Vas. 180">.
44 //. xxiii. G37 ; Od. viii. 229.

1 //. ii. 774 ; Od. iv. 626, xvii. 168.

THROWING THE JAVELIN. 267

But though in this as in all other sports the chieftain excelled the
common soldier, the javelin like the bow did not occupy a very high place in
the aristocratic sports of Homer.

To the same prejudice we may ascribe the fact that in the great athletic
festivals which preserved unchanged through all their history many of the
aristocratic traditions of their early days, the javelin and the diskos, which
as I showed in my last article may be traced back to the stone thrown in
primitive warfare, were not separate events but merely formed part of the
pentathlon. But though javelin throwing as a sport Avas less esteemed than
boxing or wrestling, the use of the javelin was universal. As tin' weapon of
the chase every Greek must from boyhood have practised throwing the
javelin for distance and at any improvised mark. At an early date its use
was taught in the gymnasia, and its popularity is shown by the numerous
representations of it on the vases and by the frequent metaphors which Pindar
borrows from it. There is however no evidence for any separate competition
in javelin throwing, with the possible exception of the competition on
horseback, until the fourth century.

The question whether in the pentathlon the javelin was thrown for
distance or at a mark has been discussed at wearisome length by archaeolo-
gists and commentators on Pindar. The argument too often revolves in a
hopeless circle, the commentators using as premisses the purely d prion
statements of early archaeologists or conclusions based on the very passages
which they are discussing. Martin Faber for example exhausts himself in
the Sisyphean task of proving that when Pindar speaks of ctkotto^ m
connexion with a javelin, he does not mean a mark but a boundary wall
evolved out of his own imagination. 4,; Others argue from conjectural
hypotheses as to the order of events in the pentathlon or the method of
deciding this competition as though the hypotheses were established tacts.
The question has been admirably discussed by Juthner, whose conclusion is, I
think, incontestable, namely that the competition in the pentathlon was one
purely for distance.

Let us first take the witness of the monuments. The vases, as we have
seen, show no evidence for throwing at a mark. The points of the javelin are
blunt, the thrower has his head turned away, and there is no sign of any
mark or target. The latter argument is particularly convincing, because on
the only three vases which represent throwing the javelin on horseback the
target is clearly depicted. It is unnecessary to discuss the delightfully naive
suggestion of M. Girard 47 that the objects falsely interpreted as compass -
which as has been shown above are really nothing more than badly drawn
amenta, were used for drawing circles in the sand which served as targets. 01
the hardly less curious suggestion of the late Mr. Freeman that these same
objects were a sort of croquet hoop used as a target. 4 " The hunter or soldier
does n>'t throw his spear at his victim"s feet but at his body, and it a targ
used it is at a reasonable height.

4,; rhllolonos, 4. 489 sq. 4: LI. ■ p. 206. p. 134.

268

E. NORMAN GARDINER

Fig. 15.— R.-F-. Kylix, Berlix 2728.
(After Jiithner. )

The literary evidence agrees with that of the monuments. The passages
of Pindar referring to a mark with the exception of the account of the games
of Heracles have no necessary connexion with any competition, certainly not with

the pentathlon. They are metaphors
borrowed from the practice of everyday
life. One passage certainly refers to the
pentathlon, 4 ' 1 two others possibly ; " J " all
three clearly indicate a distance throw.
Lastly Lucian in the Anacharsis 27 defin-
itely states irepl citeovriov /3o\?}? e'<? /jli}ko^
a/xtWwvrai. His evidence though late
is valuable because he is speaking of
Olympia, and therefore of the pentathlon,
the only event in which javelin throwing
occurred at Olympia, and it is most im-
probable that the conditions of the com-
petition were ever changed in that most
conservative of festivals.

The pentathlete then threw the

javelin for distance. As in the diskos

throw and jump he was not allowed to overstep a certain line. This line is

perhaps suggested by the pillar in Fig. 15 : it is certainly the repfia mentioned

by Pindar in Nem. vii. 70

Kv^eiuSa TruTpaOe — coyer^, uTrofxvvai

/tit) repfia 7rpo/3a<; aKovd" tore ^aXKoirdpaov opaai.

6oav yXwaaav, 09 i^iire/JL-tyev 7ra\atcr/ii('n(0i'

av)(€va teal crOevos ahiavrov, aWwvt irplv ae\l(p yvlov epareaelv.

Here I must join issue with Jiithner. He argues that the javelin
thrower with his short run would be most unlikely to overstep the mark, and
therefore concludes that the rep/na is not the line from which he throws, but
the line on either side of the arena within which he must keep his throw.
This interpretation does violence not only to the natural meaning of repfia
but also of Trpofidv, which can only mean ' stepping in front of.' Moreover,
experience shows that in the similar competitions of throwing the cricket-ball
or putting the weight, disqualification for overstepping the line though rare
is by no means unknown in the excitement of competition. With this reserva-
tion, Jiithner's interpretation of the passage may be accepted. Without fully
discussing the endless interpretations of these lines I may briefly state
my reasons.

Pindar defends himself throughout this ode against a charge brought
against him by certain Aeginetans of having transgressed the rules of courtesy
and fairness in some previous mention of their national hero Neoptolemus.
These detractors seem to have blamed Thearion for allowing one who had

; ' Nem. vii. 70.

Isthm. ii. 35, Pyth. i. 11.

THROWING THE JAVELIN. 269

insulted their hero to introduce a note of discord into the triumph of his son
(v. 69). Pindar had trangresssed the laws and was disqualified thereby.
Perhaps too they had taunted Thearion for his extravagance in employing
expensive a poet (v. 18). In reply, Pindar appeals to the fact of his friendship

with Thearion (v. Gl), to his position as proxenos at Dodona (v. GO). He-
disclaims all violence or arrogance (v. 62) ; there is nothing discordant in his
praise of Sogenes (v. 69). Then comes the passage in question. 'Sogenes of
the house of the Euxenidae I swear that I did not overstep the mark and
send forth the swift speech of my tongue like a bronze-headed javelin that
puts out of the wrestling the strong neck sweatless yet. or ever the limbs be
plunged in the sun's fire.' He disclaims all unfairness that would disqualify.
' Yet,' he continues, ' if there was trouble, if I was carried somewhat too far,
can make amends : after trouble delight follows more abundant.'
Such I take to be the thought of the passage. A word or two on points
of detail. The emphatic airoiivvw followed by fi/j surely shows that
the disclaimer is not confined to the participle 7rpo/3«? but extends to the
infinitive opaai. Professor Bury realizing this proposes to read vtto/jlvvco, a
purely arbitrary emendation of a scholiast and quite unnecessary. Secondly,
if, as Mr. Fennell says, the- notion of disgrace docs not generally attach
to iK7refnr(0, it frequently does so. It is used of 'divorcing' a wife, and
'sending into exile,' while he quotes no instance in which it means 'to
release ' or ' send off in triumph.' The word is however in itself neutral, and
takes its meaning from the context. What then is its meaning in athletics '
Did the Greeks regard the finish of a hard-fought contest as an unnecessarj
toil from which it was an advantage to escape? Every sportsman will
instinctivelv answer ' no ;' and that the Greeks really were sportsmen is shown
by the additional honour attaching to a victory in which the victor had
fought every round without drawing a bye. Certainly Pindar's ideal athlete
who • rejoices in the cost and the toil' (Isth. v. 10) would feel no satisfaction
in being 'put out of the wrestling' by an opponent's mistake.

This interpretation then does not rest on the ' arbitrary assumption '
that an unfair throw at once disqualified the competitor, it rests on the
natural meaning of the Greek. If, as I believe, the words natural ly
imply such a disqualification, it is for those who interpret them
otherwise to show that such disqualification did not take place. For my
own part, considering the punctilious, the religious strictness with which
the great games were administered, it seems in the highest d< _
probable that the slightest breach of the regulations involved disqualifica-
tion. But it is arbitrary to assume that Sogenes himself or a fellow
competitor overstept the line, it is arbitrary to assume that one competitor
frequently won three of the first four events, it is arbitrary to assume that
in consequence the competition was frequently finished before the wrestling
came on, it is arbitrary to assume (though personally I think it probable)
that the javelin throwing immediately preceded the wrestling. These and
other arbitrary assumptions are made by those who translate t^rre^ei'
TraXaicr^cnwv 'saves from the wrest lii _

270 E. NORMAN GARDINER

The competitors then might not overstep the line and any such breach
of the rules probably involved immediate disqualification. Further, common-
sense and the safety of the spectators required that they should keep
within certain limits as regards direction, and this as Jlithner sees is implied
in the e%w dycovos of Pyth. i. 44, an expression which is not synonymous
with repfxa irpoftds

eXiro/jiat
fxif \a\K07rdpaov ci/covd' oxrecT ciyco-

vo<i /3a\elv ei;(D iraXafxa hovewv,
fia/cpa Se pl-tyais dpuevaaaO' dvriovs.

How many throws were allowed we cannot say. The fact that on the
vases youths are represented frequently with two, more rarely with three
javelins in their hands, renders it probable that two or three throws were
allowed, but the evidence is not conclusive. 51 Nor do the javelins which
we see so commonly in palaestra scenes stuck in the ground allow us to
conclude that no throw counted unless the javelin stuck in the ground ;
an impossible condition with blunt points. Nor do we know how the throw
was measured. In the stadium of Epidaurus there are a number of short
square blocks facing one another on either side at fairly regular intervals
which may well have served for measuring the throw of the diskos or spear
like the measured boards on either side of the modern long jump. 52

Towards the close of the fifth century increased importance was given to
the javelin as the weapon of light-armed troops and of the Epheboi ; and
from the fourth century onwards we find d/covTMT/JLos quoted in inscriptions as
a separate competition at Athens and elsewhere. 53 The association of
the javelin and the bow suggests that in these competitions some sort
of target was used. At all events the case cited by Antiphon proves that
javelin throwing at a mark with a sharp weapon was practised in the
Gymnasia. But the only direct evidence for such a competition apart
from that on horseback is furnished by two inscriptions from Larisa of
the time of Hadrian which mention victors <tkott<Z ire^wv and (tkottw
i7T7je(ov. rA Of the details of these competitions nothing is known.

E. — Competitions on Horseback.

From an early date the javelin had been employed by horsemen both
in war and in the chase. At Athens especially horsemanship was the duty
and also the recreation of the richer classes. Plato tells us that Themistocles
himself taught his son Cleophantus not only to ride, but to throw the javelin

51 Nothing can he proved from the Cean supra, n. 21. At Athens we hear of it first
inscription which records 3 h6yxai as a prize in an inscription relating to the Thesea, B.C.
for spear-throwing, l.G. 2360 ; Ditt. Syll. 2 ii. 160-1. There is no mention of aKovriff^os
522. or aKovTHTT-qs after the commencement ol

52 TlpaKTiKa, 1902. our era.

63 Ceos, Sestos, Tralles, Sainos, Larisa: v. M Ditt. Syll. 2 ii. 670, 671.

THROWING THE JAVELIN. 271

standing on horseback and other wonderful tricks, and in the Laws lie recom-
mends javelin throwing on horseback as a useful accomplishment. 55 Some-
what earlier Xenophon in his treatise on the duties of a cavalry officer urges
the latter to encourage his men to practise javelin throwing, and to stir up
emulation among them by offering prizes. 50 They are not merely to practise
individually, but in sham fights, using iatpaipwfxeva cikovticl. In his dis-
course on Horsemanship he gives further instructions. Velocity and distance
are the most important points for war. To secure these the thrower must
advance the left side of the body and draw back the right, straightening him-
self from the thighs, and holding the javelin pointed slightly upwards. If
however the object is simply accuracy, the javelin should point straight at the
mark. At Athens there seem to have been competitions in this sport as early
as the fifth century. It is mentioned in an early fourth-century inscription,
where among the prizes for the Panathenaea five amphorae of oil are
assigned for the first prize, and one for the second tup' 'ittttov ukovti^ovti:"
In the second century a<p 'iitttov clkovti^lv occurs in inscriptions relating to
the Thesea. 5S In Thessaly, a land always famous for its horses, we find gkottw
'nnrewv mentioned in the Larisa inscription of the time of Hadrian referred
to above. 59

Fortunately we are able to supplement these scanty notices by three vases
actually representing this competition. A fifth-century aryballos from Eretria,
now at Athens, 60 a fourth-century r.-f. krater in the Louvre, 01 and a hitherto
unpublished Panathenaic amphora presented by Sir H. ( lampbell-Bannerman
to the British Museum (PI. XX). In all three the target is a shield or similar
object with a crown forming a sort of bull's-eye in the centre, raised on a post
to the level of the horses' heads. On all three vases the competitors gallop
past this target, hurling their javelins at it as they pass. On the B. M. vase
the javelins are represented roughly by a single line, on the other two vases
they have regular leaf-shaped heads : they are held a little above the shoulder,
with the point directed somewhat downwards towards the target.

On the Eretria vase the riders wear petasoi, elaborately striped chitones
secured by a belt, and high boots. The first rider has already thrown his
weapon, but his right arm is still extended to the front. His javelin is in mid
air, having missed the target, Another javelin lies broken below the target.
There is nothing on any of these vases to indicate that the competitors
started with two javelins and threw both as they passed the target. If we see
extra javelins in the field, they merely indicate that there are more compe-
titors than can be represented on the vase space. Some conventional shrubs
on this vase suggest that the sport takes place in the open country. In
the fifth century there was probably no hippodrome at Athens, and the

35 Meno, 93 d; Leg. 834 d. J. 3 ii. 671.

Hipparch. i. 6. Cf. de re eq. viii. 10. ' Collignon, 1 178.

57 iq % ii. 965. " Milliu, i. 45. Both vases ire rep

58 lb. ii. 444, 446, Ath. MiUh. xxx (1905), and discussed in P. Wolters .
p 213. Agoncn {Wurzburg I

272 E. NORMAN GARDINER

Panathenaea were held somewhere in the deme of Echelidae near the
Peiraeus.

The krater in the Louvre has a yet more festal appearance. Of the three
riders one has already thrown his javelin, the other two are about to throw.
They wear chitones like the riders on the Eretria vase, but no boots, and
instead of the petasoi they have crowns on their heads, while over them
hover two winged victories bearing crowns.

These two vases were connected by Welcker with ->) e£ "Apyovi aairi';,
but Wolters rightly refers them to rb e'</)' l-ttttov aKovri^etv. Their festal
character suggests a definite connexion with some festival, but what festival

Ipwmmmnm

Fig. 16.— Panathenaic Amphora in Britlsii Museum. B 146.

we cannot say. The sport was probably a common one in Attica, Thessaly,
and other horse-breeding lands, and formed an attractive feature of other
festivals, besides the Panathenaea and the Thesea. There is certainly no
ground for connecting the vases with the Argive Heraea.

The B.M. vase figured in Plate XX needs no detailed description. The
riders wear the regulation dress of the Athenian ephebos, a bright-bordered
chiton fastened over the left shoulder and the petasos. A similar ephebos
occurs on another Panathenaic amphora in the British Museum (Fig. 16).
Il>: carries two javelins, and beside him stands another youth naked, and on
foot, also bearing javelins. Whether this vase was a prize for the same event,
we cannot say for certain. It may have been connected with those com-

THROWING THE JAVELIN. 273

petitions tor evavhpia and evoifkia mentioned in inscriptions, of the details of
which we are completely ignorant. The immense variety of competitions at the
Panathenaea is best illustrated by the well-known Panathenaic vase repre-
senting an acrobat tumbling. ''-

There remains yet another Panathenaic vase, figured by Gerhard in his
Etruskische unci Gampanische Vaseribilder. 63 Here we see four youths
galloping to the right, but they are naked: there is no target, and their
javelins are blunt, if we can judge from the illustration. Clearly we have to
do with quite a different event, perhaps with some sort of sham fight such
as is described by Xenophon, which in later times developed into a
competition. 04

We must think of the Panathenaea as partly a vast military tournament
with a variety of displays and military competitions, which must have
appealed greatly to the spectacle loving populace of Athens. Not the least
attractive event was to e</>' 'ittttov a/covTt^etv, an event which finds its modern
analogy in such competitions as heads and posts or lemon-cutting. Of its
details and regulations we know nothing. Its popularity is shown by the
number of vases representing it ; but that it did not rank as a serious athletic
event is proved by the fact that only five amphorae were given for the first
prize and one for the second.

The javelin has been admirably treated by Dr. Juthner in his Antike
Ticrngercithe, to which I owe much. I have endeavoured to avoid covering the
same ground and to deal with points which he has discussed less fully. I
have received considerable assistance from the practical experience of Colonel
V. Balck, and also from Mr. Henry Balfour of Oxford. The vases from the
British Museum are published by kind permission of Mr. Cecil Smith from
drawings by Mr. Anderson.

E. Normax Gardiner.

62 Sal/man Xccrupolc de Camiros, PI. Comptes Rendus, 1S7»3, p. 52.
XXXVII ; Schrciber Atlas xxiv. 2. M Hipparch. i. 20 ; /.'/. 1921.

G3 i. PI. A. 3. 4 : Annal , ii. 1830, p. 223. 9;

H.S.— VOL. XXVII.

TWO NOTES ON PFLOS AND SPHACTERIA.

The Route followed by the Messenians at the Capture of the
Spartan Force on Sphactkria.

Thucydides (iv. 36) describes the last phase of the long contest which
led to the surrender of the 292 surviving Spartans on Sphacteria.

They had gradually retired to the summit of the hill at the north end
of the island, — an altitude of something under 500 feet, 1 and were making
their last stand in the neighbourhood of the 7ra\aibv epv/ua mentioned by
Thucydides, which had once defended this summit, and of which small
fragments are yet to be seen. These fragments are still there, for since
prehistoric times this practically waterless island has probably never had
inhabitants except a few nomad goatherds.

The ground, — working round west, north, east, south, — is as follows, and.
the photographs reproduced will help to make it clear (Fig. 1, and Plan)
To the S.W. is the long slope up which the Spartans had been slowly retiring
from their camp on the low level in the centre of the island. To the
west the hill falls, not very steeply, to a saddle ; and from this and all sides
except (as they thought) the east, the Spartans were exposed to attack;
then comes a shoulder before the ground slopes away westward to the open
sea. To the north a steep but easy rockstrewn descent leads to the narrow
Sikia channel dividing Sphacteria from Pylos (J.H.S. vol. xviii. Part I,
Plate IV.). To the E. of the summit there are rocks and a small cliff 30
to 40 feet high, presenting no difficulty to a climber, which drops down into
a fairly level notch (hereinafter to be called ' the notch ') some thirty yards
wide and easily visible at a distance from a northerly and southerly direction
(Figs. 1, 4, 6, 7) ; and eastward of the notch the precipice descends almost
sheer some 400 ft. to the water of the Bay of Navarino, the \i/x7jv of this
part of Thucydides.

To the south of this cliff and opening into the south end of the notch
there is a steep gully — 'the gully,'— lending down to the water, apparently
scalable in its whole length, though we only tested the upper part. South

1 Mr. Grundy's survey J.H.S. vol. xvi.) height, and their equality is confirmed by photo-
makes both Pjdos and Sphacteri; 150 feel in graphs.

TWO NOTES ON PYLOS AND SPHACTERTA.

275

again of the gully the line of cliff is quite precipitous, in places actually
overhanging (see Figs. 2, 3). But at the foot of this cliff there runs a narrow-
ledge at the top of an exceedingly steep slope of varying height above tin-
water; and along this ledge, for the most part quite close under the over-
hanging precipice (Fig. 3), a goat-track may be followed from the Panagia
Landing-place to the gully. The distance from this landing-place to the

T. - —

— /

< — '

Fig, i.— General View northwards from Cliffs South of the Panagia.

notch and summit would appear from Grundy's map (of which our Plan is
an adaptation) to be a few yards under a mile.

The data we find in Thucydides for determining the route taken by the
Messenian force are as follows :

1. They leave the Athenian main body at a point from which their
C( mniander has just been able to communicate with Cleon and Demosthi -
and

2. They reach their starting-point without attracting attention.

3. They are perhaps led by someone who has reason to think a way along
the cliff exists which will take the Spartans in rear, though he does not know

T 2

Plan of Pylos and SrHACTEiUA. (After Grumly's Survey, J. U.S. Vol. xvi.)
N.B. — The Saddle is at the point oceupie I by Idler P in. ' Position.'

TWO NOTES ON PYLOS AND SPHACTERIA.

277

it in detail. A Messenian exile from Naupactus might have .some such
reminiscences of childhood.

4. Their track was out of sight of both friends and foes upon Sphacteria :
they were only seen when they appeared upon the summit.

5. It lay along the face of the cliff wherever the ground allowed a
footing (Kara to del irapelKov tov Kprjixvwhow;).

6. The summit was gained behind the backs of the Spartans: the
Messenians when they appeared were above them.

7. The ground where the Messenians appeared was unguarded, owing
to the natural protection that the precipice was expected to afford.

Fig. 2. — Red Bluff from South.
The Messenians' startiug-point would be this side of it. Summit visible beyond (left).

With these data before us we are surely justified in drawing the inference
that it was in the notch that the Messenians gathered their forces before they
ascended to the summit.

For the route taken by the Messenian general from the time he left
the Athenian main body to the notch there are three conceivable
alternatives :

A. That they moved round the northern shore, from the west, and then
clambered up the steep but not precipitous north-east corner, reaching the
notch at its north end (Fig. 6).

B. That they took boats from the Panagia landing-place either to the
foot of the gully or to the north-east corner of the island, and tie
climbed up either to the south or to the north end of the notch.

278

W. C. COMPTON AND H. AWDRY

C. That they crept along the face of the cliff from the south till they
reached the gully, and then ascended to the south end of the notch.

To the selection of the most probable of these alternatives the following
remarks may seem to offer some guidance:

A may be dismissed : for although the ascent from the north-east
might possibly, but by no means certainly, be out of sight, yet so short is the
distance from the Spartan line of defence to the Sikia channel that the
movement of some 200 men (we can hardly suppose less were employed)

Red Bluff.

Fig, 3. — Red Bluff and ouit Route between Cliff and Bushes; looking S.

i loat-track at foot of cliff.

along this northern foot of the island to their starting-point could hardly
have passed unnoticed. Furthermore the ascent from the north-east does
iiot'in any way correspond with the difficult climbing suggested by Thucydides'
narrative.

B is open to the objection that (though various writers, including
Professor Bury, have assumed the use of boats) there is no mention or hint
of such use in Thucydides' narrative, and the whole reads like a land opera-

TWO NOTES ON PYLOS AND SPHACTERIA.

279

tion; while a flotilla of boats would be far more likely to be seen than
climbers along the cliff.

0. To prove the possibility — and to plead the probability — of this route,
we offer the following account of a scramble, successfully achieved on
April 25, 11)06, which appears to be the first recorded ascent on the part
of anyone endeavouring to trace the whole route of the Messenians from
start to finish.

After landing at the Panagia we followed at first two different routes,

Noti b.

Fig. 4.— Looking N. from almost the same point as Fig. 3.
The put farthest to the right is Pylos.

one along the shore close to the water's edge 2 ; the other ascending to the

ground above the cliff and following this in a northerly direction till within
sight of the summit and the position that would be covered by the
Athenian main force, near which the interview would have taken place
between the Messenian leader and the Athenian generals.

Here then we should have the .Messenian starting-point Fig. 2), and
they would move down over the edge of the cliff without attracting attention,
because (1) the distance from the Spartans was nearly a mile (2) the whole

'- 1 This so far as we could judge might, with a
little wading, have proved a possible track, but
the probabilities seem all in favour of the

other, and we are not in a position to assert the

ibility of the water"* eJtre i

280

W. C. COMPTON AND H. AWDRY

intervening ground was crowded with troops and (3) thick with the dust
of the burnt brushwood .

Their route led in a few minutes to a point from which the summit and
the Spartans' last stand became invisible owing to the interposition of a
bluff (Fig. 2), which falls precipitously to the bay, almost, or quite,
overhanging (see Figs. 2, 3). This bluff is a conspicuous point, being
about the last shoulder but one from the summit, and its face is more red
than the rest of the line of cliffs. 3 From this spot the Messenians' clamber
would have commenced (Plan and Fig. 2). The descent from the ridge at this
point, though fairly steep, presents no difficulty whatever. At the foot of the
cliff some 50 ft. above the water's edge our two routes joined, and from thence

Fig. 5. — Looking down from the notch, southward.
Showing top of the gully, and on the right the rocks leading to the summit.

we proceeded together practically all the way to the top, using a goat-track
(Fig. 3) that rose and fell keeping close under the main cliff, the general
trend of the track being upwards, so that we gradually rose to a considerable
height above the water, having on our right all the way a full view of the
harbour and the sand-bar with the lagoon beyond it, and occasionally
catching sight of the notch (Fig. 4;. 1 Here and there a ledge had to be
traversed requiring hand- as well as foot-hold; but both were always forth-
coming, so that there was no point of difficulty for anyone used to rough
country. After going for about half an hour we saw before us the gully
mentioned by Professor Burrows (J.H.S. vol. xvi. Part 1,1896) which was our

It can easily be identified also in the photograph taken fiom Voithio-Kilia (Fig. 7).

TWO NOTES ON PYLOS AND SPHACTERIA. 281

objective. Professor Burrows assumes that the Messenians reached tliefootof
this gully by boat and ascended it all the way to the notch. It has however
been shown that our track fulfils the requirements of the narrative, in which
no mention is made of the water. We had started from a point where tie-
Athenian generals could be conferred with, the whole route was completely
out of sight, the cliffs rising vertically above us to a height of 200 ft. or more,
and our track following the cliff as best we could find a footing (kutci to
del Trapelicov). No better cover could be desired (see the illustrations).

On reaching the gully in about three quarters of an hour from the start,
we proceeded straight up it till our way was closed by a projecting buttress
only allowing of an ascent by a chimney, which involved the use of the back
and knee, for 12 or 15 ft. After this point, which one of us avoided by
descending to the main gully, — a wider hollow to the north of the buttress, —
there was no difficulty, except the thickness of the bushes, in proceeding
upwards to the notch. The head of the gully was reached in something less
than 1| hour from the start. From the notch to the summit, as has been
shown, the final scramble of the Messenians would be accomplished in a
very few minutes; so that we may conclude that they were sighted on
the summit within 11 hour of the time when they offered to the Athenian
generals the prospect of seeing the Spartans outflanked.

Incidentally also another point comes out. We were of course', in full
view of everything to the east and north-east of us; and had the Spartans
been in possession of any part of the sand-spit they would have been near
enough to have found some means of attracting their friends' attention to
the Messenians' movement, which clearly they were not. They must there-
fore have had their camp at a distance away to the east or north-east of the
lagoon. However, this remark applies equally to whichever of the three
alternative routes we select.

W. C. COMPTON.
H. AWDRY.

II.

The Land Defence of Pylos (see Plan, p. 270).

I had greatly wished also to investigate thoroughly the competing lines
of the Athenian defence of Pylos on the land side, but time ran short before
I had by any means satisfied myself; indeed, if I express an opinion at all.
it must be with most humble apologies to those of much greater local and
archaeological knowledge than myself

In a former article on the strategy at Pylos and Sphacteria J.H.S. vol.
xx. 1900] I ventured to point out that Demosthenes' line of defence would
have been governed by the following three military considerations : (1) the
least possible amount of building to be done. 2 the fortress to be reduced
in size as far as possible. (3) all dangerous or doubtful ground to be left
outside the fortress.

282

W. C. COMPTON AND H. AWDRY

The competing lines are two : (^l) that favoured by Mr. Grundy and
shown in his map (J.ff.S. vol. xvi, of which the Plan in this article is an
adaptation), viz. a line running north and south along the sand-hills in
continuation of the east cliff of Pylos northward to Voithio-Kilia ; (B) that
favoured by Professor Burrows, a line running east and west along the
high part of the] north cliff of Pylos, from Nestor's cave westwards, with a
wall to continue it from the western end of this cliff to the north-west
corner of Pylos peninsula, where the western precipices overhanging the
sea become high.

Of this (A) gives a strong line, for the sand-hills are very steep and sand-
climbing is very difficult (Fig. 6), and the piece where a wall would have to
be made is short ; but it leaves much more ground on Pylos for the defenders

Nestor's Cave.

JNestor's
"(Cave.

Fig. 6. —View of Pylos and Sphacteria from the X.
The photograph illustrates the epithet 'Sandy' as applied to Pylos.

to hold, including a very considerably longer piece of coast. This coast is for
the most part low, but sheer, precipice ; but I saw one spot, — and there may
have been more, for I had not time to go all round, — at the promontory form-
ing the south side of the entrance to Voithio-Kilia (Fig. 7), where a ship
or two could have landed men if unopposed, and these once landed could
with ropes or otherwise have helped their comrades up elsewhere. Demo-
sthenes therefore if he held line (A) must certainly have spared men from his
small force to watch this coast, a necessity he would of course wish to avoid.

As to line (2>), the northern cliff is really impregnable, and gaps in it
have been at some time strengthened by fortifications rough and otherwise.
The chief piece of fortification here is of so regular a character that it cannot
be part of Demosthenes' hasty building, and would probably belong, like the
extensive Greek foundations on which the neighbouring mediaeval castle on

TWO NOTES ON PYLOS AND SPHACTERIA.

283

the summit is built, to the permanent Athenian fort held by them from
B.C. 425 to 413. The only vulnerable point is the north-west coiner between
the north cliff and the sea; and without discussing the question whether
the still remaining wall near that point is the Athenians' work or not, there
would have been no difficulty in the time at Demosthenes' disposal in
supplying an adequate wall. It would therefore seem likely that Demo-
sthenes would be adhering best to military principles in taking the line (B) ;
and that although this line gave him perhaps slightly more building, he would

I J

(Landing

Fig. 7.— View of Pylos \ni> Sphacteria from the X. boundary of Voithio-Kilia.
Nestor's cave may be distinguished here also.

prefer to contract his fortress, economise men, and leave the low sea cliffs and
the low ground at the northern end of Pylos outside his fortress.

We must remember that in speaking of Pylos we are dealing with
a place where, unlike Sphacteria, there has been much fighting subsequently
to Demosthenes: (1) in the 12 years of the Athenian occupation, (2 in the
Middle Ages as evidenced by the castle on the summit and the tower at the
south-east corner, (3) in the War of Independence; and that there are the
remains of buildings of all manner of periods upon the peninsula.

To what period do the numerous skeletons sticking out of the ground
at the southern end of Pylos over the Sikia Channel belong ?

II. Awdrt.

ARCHAEOLOGY IN GREECE.

(1906—1907.)

The archaeological activities of the twelvemonth from July 1906 to
June 1907, with which this paper deals, have been very numerous. Not only
are there many smaller discoveries of interest to be recorded, but also the
more important excavations, the French at Delos, the German at Pergamon,
to which may now be added those of the British School at Sparta, have made
considerable progress. Especially important is the scheme for excavating the
Agora and the northern slopes of the Acropolis at Athens, now at last begun
after many difficulties by the Greek Archaeological Societ}^.

The work is under the direct supervision of the General Ephor of
Antiquities, Dr. Kavvadhias, with a committee consisting of the Crown
Prince as chairman, the president of the University, the mayor of Athens
and the directors of the foreign Schools of Archaeology. A beginning has
been made by the demolition of some houses near the Theseum. The
scheme embraces the clearing of the whole of the northern slopes of the
Acropolis, the Agora, and in general, the region between the Tower of the
Winds and the Theseum. This area is entirely covered with houses, and
this has given rise to serious difficulties. These are now in a fair way to be
overcome, and great results are to be expected in the course of the next few
years.

Another interesting excavation at Athens was in progress in August of
this year at the Dipylon Gate under the direction of Dr. Bruckner and
M. Skias. They have proved that all the monuments were in family groups,
and stood, not level with the road, but on built platforms some nine feet
high. Each family had its own platform. There is also evidence for dating
all tin' monuments to the period between the year 393, the date of the
Dexileos Stele, and 317 B.C.

Perhaps the most important excavation of the Society is that conducted
in the autumn of 1906 by Dr. Stais at Sunium. Whilst digging near the
temple, in the hope of finding relics of the period before the Persian Wars, he
hit upon a cleft in the artificially smoothed platform of rock, upon which the
temple stands. In this he found two colossal archaic male statues of the
'Apollo' type, with the base and feet of a third, and the base and feet of a
fourth figure, that was possibly female. They were associated with
Corinthian pottery. One of the ' Apollos ' has lost its head, and has been
left, with the base, at Sunium. The other is now in the centre of the

ARCHAEOLOGY IN GREECE. 285

Archaic Room in the National Museum at Athens, where its huge stature —
it is eleven feet high — dwarfs all the other ' Apollos' amongst which it stands.
Iu style it belongs to the more advanced class, with one foot forward, and
the arms nearly free. It is complete, but for the shins and part of the face.
These have now been restored with plaster, unfortunately of so nearly the
same brown as the statue itself, that it is difficult to sec exactly what is old
and what new.

Its original position is proved by the discovery on the rock-platform
outside the temple of a hewn square depression, that exactly fits the base.
All four statues were probably dedications set up outside the temple, where
their size must have made them conspicuous objects from the sea. Since
the Persian invasion they have lain in the cleft in which they were found. 1

Dr. Klon Stephanos has continued his researches in the prehistoric
cemeteries of Naxos and Syros. These are all of the Cycladic period, with
the skeletons lying on the side and the legs drawn up, and have yielded a
great number of characteristic objects, marble female idols, obsidian knives
and hand-made pottery, either incised or, later, painted with geometrical
patterns. Dr. Stephanos is of opinion that the relative age of the tombs is
indicated by this, but still more by the greater or less quantity of bronze
present.

These cemeteries, which have now been found in so many of the ( velades,
suggest some problems that have as yet hardly been answered. Where are
the corresponding settlements, such as that of which scanty remains were
found at the lowest levels at Phylakopi, and why are the remains between
this period and the Mycenean age so scanty ( These Cycladic cemeteries
with their 'Amorgine' culture are contemporary with Early Minoan III in
the Cretan scheme, and abound all over the Cyclades. For the succeediii-
Middle Minoan and Late Minoan I and II periods we have in the Cyclades
only Phylakopi in Melos, some vases from Paros, and the prehistoric finds in
Thera. For the Mycenean period (Late Minoan III) the case is almost the
same: very little has come from the islands. It would seem as if there were
three periods in the Bronze Age. In the first, that of the 'Cycladic" or
'Amorgine' cemeteries — the Early Minoan of Crete — the Cyclades held the
first place in the Aegean world, and largely influenced Crete. 2 In the second
they gave way to Crete, the highest point of whose culture Dr. Evans puts
in Middle Minoan IIT. In the third and last period the mainland of (ii
took the lead. Only all through the obsidian trade enabled Melos to hold a
good position. It is much to be hoped that some other prehistoric settlement
like Phylakopi will be excavated, to give in the light of present knowledge a
continuous picture of Bronze Age culture in the islands.

At the Amphiareum at Oropos ML Leonardos has found a fountain and
the remains of houses and shops that were used for the accommodation of
pilgrims to the shrine.

1 Ath. Mittk. 1906, i». 363. To lie fully Cyckdic style uear Chalkis and on tl

published in the 'E<py),j.(p\s 'Apx at0 ^°y'K'h-

2 Sc* below, the notices of tombs "f the

286 R. M. DAWKINS

Near Chalkis, M. Papavasileiou lias excavated a number of tombs, some
Mycenean and some said to resemble the Cyeladic burials. Many vases are
reported. The Mycenean tombs are at a place called Vromoiisa, and belong to
the very end of the period. One tomb even contained a geometric vase.
The 'Cyeladic' tombs are at Manika, about an hour from Chalkis, and consist
of the tomb proper, covered with slabs, and a small dromes. Two marble
idols were found, one female, and badly baked pottery. I have not seen the
objects.

Besides continuing his work in Lokris and at Thermon in Aetolia, Dr
Sotiriadhis has excavated neolithic sites at Chaeronea and Drachmani. No
metal was found, but celts, obsidian, and hand-made pottery. Of this some is
black polished ware ornamented with incised patterns, and some has geometric
patterns in red paint, and resembles the very remarkable pottery found in
the neolithic settlements at Sesklo and Dimini in Thessaly. This northern
Greek neolithic seems to have no connexion with the Aegean culture of Crete
and southern Greece, but to have its allies rather in the north and west.
The pottery has resemblances to that found by Dr. Doerpfeld in Leukas,
and its context seems to be far more European than Aegean. The celts from
Thessaly, it is noticeable, have squared edges, and are entirely different from
those so common in southern Greece and Crete. The special importance of
an examination and comparison of these remains, for the most part still unfor-
tunately unpublished, appears below in the discussion of Dr. Doerpfeld's
recent discoveries.

The work of M. Skias at Corinth will be mentioned in connexion with
the American excavation.

An interesting discovery is announced from Thebes. In the middle of
the modern town, which occupies the site of the Kadmeia, a Mycenean house
has been accidentally discovered, and partly excavated by M. Keramopoulos.
It contained fragments of wall paintings, one shewing a forehead with curls,
another a face, and others parts of the hair. Besides these, which formed
part of figure subjects, is a piece shewing a lily. There were also five very
large pseudamphorae, as much as two feet high, and two hundred or more
cups and Mycenean kylikes, many absolutely new, as if from a shop. The
house had been destroyed by fire. The site is to be expropriated, and fully
cleared. Mycenean remains have been found before at Thebes, but nothing
so promising as this house, which from its central situation and fine paintings
seems likely to be a palace or building of some importance.

Great activity has been shewn this year in building and enlarging
Museums, and for this work, which is more useful and necessary than
showy, the Archaeological Society deserves great credit and the thanks of
all archaeologists. The General Ephor, Dr. Kavvadhias, recognizes that it is
not only necessary to dig up antiquities, but also to preserve and exhibit
them adequately, and all students reap the benefits of this wise policy. The
general tendency is now towards building local Museums, and, although this
involves the visitor in a good deal of travelling, it is often an advantage to
see the objects in the place where they were found.

ARCHAEOLOGY IN GREECE. 287

Thus at Epidauros Dr. Kavvadhias lias, besides conducting a supple-
mentary excavation, arranged the Museum, and taken measures for the safety
of the architectural remains. Study of the architectural fragments has
also enabled him to erect in the Museum reconstructed parts of the more
important buildings.

Other work of this kind has been undertaken at the tempi'' of Bassae,
where the walls have been almost entirely rebuilt from old material, and the
half-columns inside the temple re-erected.

At Corinth, the old temple and Peirene are being put into a safe
condition.

Nor have medieval buildings been neglected, and the Greeks no longer
deserve the reproach that they are careless of the monuments of their later
history. In particular, the Byzantine churches at Mistra, which were in a
very dangerous state, are being repaired under the care of M. Adhamandiou.

A new museum has been built at Lycosura, to exhibit the colossal group
of Despoina, Demeter, Artemis, and Anytus by Damophon. The actual
work of re-erecting the statues is now in hand, and casts of the pieces
in Athens have been sent to Lycosura. The work has been undertaken by
M. Kourouniotis, and the sculptures have been the object of a special study
by Mr. Dickins of the British School, who has the permission of the
authorities to publish the reconstructed group. M. Kourouniotis has also
excavated round the temple, finding a part of the wall of the sacred
enclosure.

The enlargement of the Museum at Sparta is mentioned below.

A museum has been built at Corcyra, near the tomb of Meneci
and the programme of the Society includes the building of several mere of
these local Museums, which add so much to the pleasure of travel in Greece,
and have moreover the advantage that objects in themselves of minor
importance are not lost, as they are apt to be, in a very large central
Museum. 3

This year was the second season of the work of the British School at
Sparta. The further excavation of the Sanctuary of Artemis Ortbia and of
the city wall coutinued work begun in 190G. New ground has been brok< a
by the discovery of the site of the temple of Athena Chalk ioikos.

The main objective was the Sanctuary of Artemis Orthia. The buildings
at this site are a temple built probably in the sixth century B.C., and lasting
en until the third century A.D., although rebuilt during the Hellenistic
period. Secondly, a Roman theatre, built at the end of the second or begin-
ning of the third century A.D., in which tin' facade of the temple was included,
occupying the position of the stage building. The Roman theatre has new
been completely cleared. In the anna or orchestral area were found the
remains of the altar, built at the same Roman period as the theatre itself.
Beneath this altar were blocks that belonged to the altar of Hellenistic times.

These not.-s have been largely drawn up from the UpaKTixa. of 1'.' 6 logical

Society.

288 R. M. DAWKINS

and in connexion with them a deposit of burnt refuse from sacrifices and
some late Greek sherds and terra-cottas.

More than a metre below the Hellenistic level a deposit of archaic Greek
objects was reached : this has now been cleared down to solid earth all over
the arena and inside the temple. Above the archaic deposit was a layer of
sand which had been brought from the river to raise the level when the
temple was built — probably, to judge from the objects found in the sand,
about the middle of the sixth century B.C. The deposit below the sand is in
parts as much as a metre thick, and ranges in time from the eighth, or
possibly the ninth, century to the middle of the sixth century B.C. Very
near the bottom of this structure is a cobble pavement, on which stands
a large altar built of stones in regular courses. This altar is directly
below the Hellenistic and Roman altars. The temple that existed con-
temporaneously with it has not yet been found, but there are indications
that its remains are below the foundations of the Roman building. This
archaic altar was surrounded by a mass of burnt matter, amongst which were
a quantity of fragments of burnt bones. The surrounding deposit contained
a great number of small objects and pottery. It was dug in layers, with the
result that at the lowest levels no pottery except ' Geometric ' was found ;
above this, ' Geometric ' mixed with ' Protocorinthian ' and a ware akin
to ' Corinthian,' whilst at the highest levels nothing but this last kind
occurred. With the pottery were found a large number of small bronzes,
pins, fibulae, and animals, lead figurines, and carved ivories. These latter
were either small figures of animals or men in the round, seals with devices
cut in intaglio, or plaques with scenes carved on them in relief. Many, if
not all, of these plaques were fastened by bronze rivets on to the front of
fibulae. The subjects represented on them comprise male or female winged
figures grasping birds, a warrior stabbing a gorgon, a dead man on a bier,
a ship with full rigging and crew, sphinxes, a man on horseback, and others.
Jewellery, engraved gems, terra-cotta figurines, some representing probably
the image of the goddess, fragments of terra-cotta masks, and other objects
were also found. The occurrence of amber, in view of the northern origin of
the Dorians and of its rarity on classical sites, is of great interest.

Thus the cult of Orthia began in the earliest times with a large altar
and probably a temple. This altar was covered up when the temple corre-
sponding to it was destroyed in the sixth century, and a new temple built
a little way off, the level being at the same time raised by the layer of sand
mentioned above. In Hellenistic times this temple was rebuilt, but lasted on
on the same site until the end of paganism. Under the late empire it was
surrounded by a theatre, from which the rites performed in front of it could
be conveniently witnessed. The altar always was in the same place, which
it occupied with ever-rising level for at least 1100 years. Except one stray
lentoid gem, nothing whatever of the Mycenean period has been found.

The sanctuary of Athena (Jhalkioikos was found behind the theatre on
the Acropolis Hill. A mass of geometric pottery shews that this sanctuary
also goes back to a very early period. The building itself was much destroyed,

ARCHAEOLOGY IN GREECE.

bat the finds were important. A very tine Panathenaic amphora, bronze
statuettes, and a large archaic inscription were found, which prove
continue, though not to complete, the well-known ' Damonon ' inscription.

The work of tracing the course of the ancient city wall was continued.
This has again been done largely by the discovery of tiles stamped with
the information that they were public tiles used for the walls. The name
of the tyrant Nabis found on some of them connects the building of the wall
with him. In a few places the actual wall has been found with remains of
towers.

In looking for the Agora some Hellenistic tombs were found, well built
of ashlar, and containing vases and discs of stout gold-leaf chased with
patterns of wreaths and flying birds. These bracteates, some of which are
double, are imitations of Sicyonian coins.

It is proposed next year to continue the work of the Orthia Sanctuary
by removing some of the Roman foundations, and thus getting at the archaic
deposit below them. In especial it is hoped that the temple connected with
the archaic altar will be found. There is also more work to be done in
digging a mass of ' Geometric ' deposit at the Chalkioikos site.

An excavation at the site usually known as Cape Sepias on the Mag-
nesian peninsula resulted in the discovery of a church with a fine mosaic
pavement, ;md some tombs of the 'Geometric' period. The vases found in
them resemble the Geometric ware of northern Thessaly, of which specimens
have been found in the Islands and in Crete.

The year's work in Crete shews that the island is by no means exhausted.
The excavation by the British School at Palaikastro and that by the University
of Pennsylvania at Gournia are finished, but fresh discoveries of the greatest
interest still continue to be made. Dr. Evans at Knosos, Dr. Xanthoudhidhis
at the Early Minoan Settlement of Koumasa in the Messara plain, and Mr.
Seager at a new site on the island of Pseira, have all done work as important
is any that has preceded, and at all three sites work is to be continued for at
least another year. The greatest promise for the future is still at Kn
where Dr. Evans has shewn that much still remains to be done, before it will be
possible to regard the Palace as fully excavated.

A building with remains dating from Middle Minoan III, the period
regarded by Dr. Evans as the highwatermark of the Minoan civilization, has been
found underneath the pavement of the West Court, and a close examinal
of the already excavated parts of the Palace has yielded remarkable results.
These have been so fully described by Dr. Evans himself in the !•
to the 7'///os i from which these notes are taken that it is not necessary h<
to do more than briefly note the more important. The restoration of
the wall-painting in the miniature style representing a Minoan temple has b
completed, and the rains of several winters have so washed the Palace
walls that "ii the west Facade of the Central Palace Court marks of
the bases of two pairs of small columns have been observed. These fit

•» Appearing in the Times I '/'■ t July 19th, 1907.

H.S. — VOL. XXVI f.

290 R. M. DAWKINS

the columns on the painting so well that, combined with other indications,
they shew that here was the facade of such a building as is shewn in the
painting.

Almost equally striking is the discovery of the actual foundation
of the stepway that led up to the balls above the basement rooms of the west
wing. A large south-western quarter of the Palace hitherto unsuspected has
been found, and a huge beehive chamber cut in the rock. This is probably
a great tomb, and its exploration is as full of promise as anything else on this
truly wonderful site.

Dr. Luigi Pernier of the Italian Archaeological Mission has kindly
furnished notes on the work this spring at Phaistos and Prinia.

At Phaistos the excavation of the earlier palace, remains of which exist
below the floors of the later building, has yielded good results. A square
room has been found, and in the middle of it a round cavity containing
lamps, Kamares pottery, and burnt bones. The latter are considered by the
excavators to prove it to be a sacrificial pit.

The position of the remains of the different periods one above another
has been examined by means of a special trial-pit. The foundations of the
earlier (Middle Minoan) palace were found to go down to about twelve feet
below the pavement of the later palace. Below these foundations were found
walls made of small stones belonging to a still earlier building, which in its
turn rests on the Neolithic deposit. The general resemblance to the results at
Knosos is striking. At both sites there are the two palaces, and underneath
them the thick layer of Neolithic remains.

Work has also been done on the later palace, where a careful examin-
ation of the walls has allowed conclusions to be drawn as to the size and
disposition of the now destroyed upper story. The area on the highest part
of the acropolis which, from the discovery of two columns, had been already
called the perisbilio, has been cleared, and now justifies its name. A square
court has been revealed, no doubt open to the sky, surrounded by twelve
columns, four on each side, with a portico running all round. In the hot but
windy climate of Crete such a cloistered court must have been a great feature
in a building, providing shelter from sun and wind alike.

Later in the season the excavators resumed work at Prinia, an elevated
site that overlooks the road from Candia to Gortyn. Some years ago
prehistoric clay idols, similar to those from Gournia and those mentioned below
from Koumasa, were found here. Now archaic Greek remains have been
discovered, notably the ruins of a temple. Its importance is attested by the
sculptures that adorned it. Part of the sima, nearly three feet high, has
been found, decorated in low relief with a procession of riders armed with
round shields and lances. This recalls the terracotta sima of the temple at
Palaikastro with its reliefs of chariots and warriors, who are similarly armed. 5
In both cases also traces of colour have been found. Besides these, there are
numerous fragments of statues in soft stone. One of the best is part of a

' B.S.A. xi, PI. XV.

ARCHAEOLOGY IN GREECE. 291

statue of a female divinity seated on a high-backed throne, decorated with
figures in low relief. To a larger, but similar figure, half life-size, belong
part of a very fine head, waving hair, and part of the breast. Associated
with the fragments of the sima were remains of some very fine pzthoi,
adorned with various figures in relief. Notable amongst these is the
standing full-far-e figure of a winged goddess, between °a pair of horses.
The scarcity of Greek remains in Crete makes these discoveries all the more
interesting. It is in the works of this period that the influence of
Dipoenus and Skyllis is to be locked for, and it is greatly to be hoped that
the work at Prinia will yield still more in the future.

The most important work on a new site has been that of Mr. Seager, who
has continued the excavation he began in 190G on the small island of Pseira,
which lies about two miles off the north coast of Crete in the gulf of Mira-
bello, off the modern village of Kavousi. That so small an island, only a mile
and a half long by one wide, and now quite uninhabited, should have been
the site of a prehistoric town, is eloquent for the populous condition of
Minoan Crete. The town, whose existence was proved in the beginning of the
campaign of 190G, lies mainly on the south-east side of the island, where a
long rocky point forms a good harbour for small boats. From the old landing-
place in this harbour a long flight of stone steps leads to the summit of the
point, and there divides into four roads intersecting the top of the hill. Owing to
the abundance of stone at hand, no brick seems to have been used in the upper
parts of the walls, as was the case at Palaikastro, and the houses were built
throughout of roughly hewn stones. The result of this solid construction is
that some of the walls still stand to the height of nine feet, clearly shewing the
thresholds and floor-levels of the upper story. On the other hand, the
quantity of fallen stones has made the work very laborious, and broken the
very fine pottery rather badly. Potsherds, found in crevices of the rock, prove
that the founding of the town goes back to the Early Minoan period, and in
fact, underlying the existing houses, were well preserved Middle Minoan I
house-walls. Directly below the floors of the houses are deposits dating to
Middle Minoan III, and the objects found in the houses themselves are Late
Minoan I and IT. Between these two styles no dividing line in the stratifica-
tion of the site can be discovered and Mr. Seager is inclined to think that tin-
Late Minoan II (Palace Style) objects may be importations. Of anything later
than Late Minoan II not a trace has been found, and it would seem that the
island town, which could only have been inhabited when the Cretans had
command of the sea, was deserted after the catastrophe which involved the
destruction of the Palace of Knosos. On the other hand the town of Gournia
on the mainland of Crete near by lasted on until the close of the Bronze A^e.
Many of the finds are very important. Besides other fragments, a
painted gesso duro relief of a lady in a wry richly embroidered dress was
found. Of this work, which must hive been about a yard high, the bust and
part of the skirt are well preserved. The best known example of such
painted reliefs in hard plaster is the figure wearing a lily crown from Ki,
That such works should be found here shews that Mr. Seaeer is rierht in

-1

-J92 R. M. DAWKINS

supposing that it was a richer settlement than Gournia. No palace has been
found, but one very fine house is built on three terraces rising from the sea to
the top of the hill. Of Late Minoan I pottery, large jars painted in white are
reported, one very fine example having three large bull's-heads on the shoulder,
with double axes between their horns, and rows of double axes round the rim
and foot, with bands of spirals filled with flowers. To the two painted clay
bulls that were found last year, a third example has now been added, and
parts of six in all have now been found. The rest of the pottery requires
mending before any more can be said of it. The richest finds were
perhaps the stone vases, of which some sixty have been found, comprising
lamps and bowls of different patterns. Another season is required to finish the
excavation, and in the meanwhile trials are being made on the neighbouring
island of Moklos.

A later notice sent me by Mr. Seager reports the discovery of an early
cemetery on Pseira. About thirty-three tombs have been opened, and more
remain. They all date to Early Minoan II and III and Middle Minoan I,
and are of two types : rock-shelter burials, common in east Crete, where the
body is laid under an overhanging ledge of rock, and cist-graves. The cist is
formed of stone slabs for the bottom and walls, and the resulting box
covered with large flat stones. About 100 vases of terracotta, and 90 of
stone are reported. The latter are especially fine, and interesting both for
their shapes and their materials, being made of breccia of all sorts, alabaster,
rock crystal, and finely coloured stones. The presence of cist-graves is a point
of great interest, as this is the type of grave that characterizes the con-
temporary ' Amorgine ' cemeteries of the Cyclades. It has been recognized
for some time that at the end of the Early Minoan period there was a strong
wave of Cycladic influence in Crete, but this is the first time that actual
graves of the Cycladic type have been found. The Early Minoan burials
found elsewhere have been tholoi as at Phaistos and Koumasa, complexes of
walled compartments, like rooms in a house, as at Palaikastro and Gournia,
or burials in rock- shelters.

In July of 1906 Dr. Xanthoudhidhis continued his work at the pre-
historic site of Koumasa in the central plain of Crete, ten kilometres south
of Gortyn. At this site he has previously excavated three circular ossuaries,
which contained bones and numerous Early Minoan III remains, notably
ivory seals with geometric designs, triangular daggers, and objects that
suggest a strong Cycladic influence. Outside these tombs he has now found
a kind of courtyard, in which was a great mass of charcoal and half-burned
bones. These he takes to be human, and draws the conclusion that before
burial the bodies were partially cremated outside the tomb proper. This
observation is of the first importance, for it has always been supposed
hitherto that cremation was altogether unknown in the Bronze A^e in
Greece. Associated with these remains were found Cycladic and Kamares
(Pearly Minoan III and Middle Minoan) pottery, and stone and ivory seals.

At the same time an examination was made of the settlement on the
north and west slopes of the hill above the tombs, which proved to be of the

ARCHAEOLOGY IN GREECE. 293

same period. The houses, built cf undressed stones, were much destroyed,
but one at least shewed traces of a second story. On the top of the hill was
the shrine, and this at all events continued in use until the Late Minoan III
period, since it contained terracotta idols like those found at Gournia and
Prima.

The shrine itself consisted of several small compartments, the most
important of which was paved, and had a central column supporting the
roof. In this were found two aniconic idols of clay, a cone and a cylinder,
and by them a steatite table of offerings, with a hemispherical hollow above,
exactly like the one found in the Middle Minoan shrine in the palace of
Phaistos, together with a sacred baitylos. The association of these sacral
objects with a pillared room is yet one more piece of evidence for the religious
character of these central pillars, found in so many Aegean buildings. There
are excellent examples at Phylakopi and Knosos. 6

In another compartment were the terracotta idols mentioned above.

At Haghia Eirene, half an hour east of Koumasa, are traces of another
contemporary settlement, and close to it Dr. Xanthoudhidhis has excavate* I
two more tombs like those of Koumasa, but robbed and re-used in Mycenean
(Late Minoan III) times. A large ossuary was examined at Porti, four to
five kilometres north-west of Koumasa. The internal diameter of the tlwlos
is 2o feet ( 7 metres), and the layer of bones three feet thick may represent
as many as a thousand bodies. Here again were the same signs of cremation.
The objects resemble those from Koumasa. Traces of other tombs were
found and of the settlement belonging to them. Work is being continued
at this most important site in the July of this year. 7

With all these fruitful excavations going on, the Candia Museum grows
steadily richer. A new Museum is being built on the eastern part of the
Venetian fortification of the town, near the Treis Kamares Square.

The French School is continuing its great task of the excavation of
Delos, and much progress has been made in clearing the town and public
buildings. The most interesting discovery is that of Mycenean remains,
carrying the history of Delos back into prehistoric times. These consist
of an ossuary by the Colonnade of Antigonus, near the Apollo Temple.
It is built without mortar, and surrounded by a wall of Hellenistic date.
Adjacent to the wall is an oblong platform. In the ossuary were Mycenean
vases, including pseudamphorae, and some two-handled spouted jars of a
type well-known in Crete, that goes back very much earlier. It is most
probable that this difference is accounted for by the re-use of the ossuary.
Another similar enclosure has been found inscribed aftarov, which leaves little
doubt that this was also an abaton, a sacred enclosure over the tomb of a
hero of the prehistoric age. It is yet another case of that continuity of a
sacred site from Mycenean times onwards, which occurs in so many other
places, and is of so much importance for the history of Greek religion.

6 For this subject, see Evans, Tree and Pillar ' These notes are largely indebted to nar.

Worship. aO-nvaia. Oct. 15th, IS

294 R. M. DAWKINS

A most interesting paper was read at the French School this spring,
the object of which was to identify an archaic marble lion, now in front of
the Arsenal at Venice, with one of the series of such figures now at Delos.
The very convincing argument rested on the similarity of the statues, and
the fact that, when Tournefort visited the island, he saw a lion which
certainly is no longer at Delos. To this we have to add the natural fondness
of Venetians for statues of lions, as being the badge of their patron St. Mark.

The work of the American School at the site of ancient Corinth was
broken off for a year, owing to the untimely death at Athens of the late
director, Dr. Heerinance. It has now been resumed under the new director,
Mr. B. H. Hill, who has kindly sent some notes of the work this spring.

As before, the depth of soil to be removed has been very great, and the
fact that the Roman level is very close above the Greek has been against
the discovery of many remains of the earlier period. Among the finds are
inscriptions, a considerable part of an ornate circular building, dedicated
by Babbius Philinus, and four headless, though otherwise complete, statues
of good Roman workmanship. The Odeum has been located, halfway between
the Theatre and the Fountain of Glauke, exactly in the position given by
Pausanias. It has a diameter of about 98 yards (SO metres), and is partly
cut in the solid rock and partly built of opus incertum.

The greater part of the work has been devoted to clearing the Roman
shops north of the western part of the Agora, and what was still covered
of the long Greek stoa north of these shops. It was here that the depth
was very great, and some half a dozen systems of walls, modern, medieval,
and Byzantine, had to be removed before the Roman level was reached. In
the face of these difficulties the excavators deserve great credit for their
perseverance, and their efforts have now removed all doubt as to the identifi-
cation of the old temple as that of Apollo, and of the sites of the Fountain
of Glauke and the Agora. Mr. Skias, however, who has also dug at Corinth,
on behalf of the Greek Archaeological Society, would put the Agora much
further east than the American excavators think possible.

Prehistoric objects also are not lacking. Near the south edge of the
hill, on which the temple of Apollo stands, fragments of prehistoric pottery
have been found, together with stone age implements of obsidian and flint,
and the torso of a primitive nude female statuette of marble about six inches
high. The obsidian was found close to the native rock, and the statuette
about a foot higher up. It will be of great interest to know whether this
fiud belongs to the neolithic series of Thessaly and Boeotia or not.

The old temple has been strengthened by the Greek Archaeological
Society, and its appearance improved by the removal of an unfinished school
built by Kapodistrias, and by the complete clearing of the foundations.

Besides the great work at Pergamon the German Institute has made
several smaller but extremely interesting excavations, at Tiryns, Olympia,
and Pylos, besides \)v. Doerpfeld's work at Leukns/

8 These notes are taken from ivports published in Aih. Mitth. xxxii, pp. i xvi

ARCHAEOLOGY IN GREECE.

At Tiryns the lower strata of the Palace were examined, with thi -
excellent results that always attend the dissection of a prehistoric site..
lower citadel was trenched, and search was made for tombs.

By making trial-pits through the floors of the palace, remains were found
of an older fortress and palace. Already in 1905 an earlier gate was
discovered below the great Propylaeum,' 1 and this has now been cleared.
The walls, preserved to a height of as much as nine feet, are built of
large stones. It was also found that the walls of the fortress are of different
dates. In particular, the eastern part of the wall, the galleries, .aid the greot
tower in the south wall are proved to be later constructions dating from the
period of the second palace.

The great antiquity of the site is shewn by the discovery, below the floor
of the earlier palace, of graves, and again below these of two strata of remains
of walls. Similar results were obtained in the middle fortress. The conclusion
of this examination of the lower strata at Tiryns ought to go a long
way in shewing us the condition of the mainland of Greece in pre-mycenean
times.

The post-mycenean history of Tiryns has also been illustrated by
the discovery of a thick layer of terracotta figures outside the south-east
corner of the upper citadel. These come apparently from a sanctuary of Hera,
and represent a seated goddess and her worshippers bringing gifts. Similar
figures were found in the Megaron of the upper citadel, and it is probable that
they all come from the Temple of Hera, that occupied the site after the
destruction of the Mycenean palace.

A number of graves of the ' Geometric ' period, generally small built square
structures, have been found between the citadel and the railway station.
The excavation is to be continued next year.

At Olympia the work at the Temple of Hera and the Pelopion begun in
1906 has been continued. 10 More trial-pits were sunk below the opisthodomos
and cella of the temple. Again sherds were found of the peculiar kind that
Dr. Doerpfeld has found at Leukas, and now also at Pylos in the excavation
mentioned below, and regards as the earliest Achaean pottery. Holding
that the culture of Mycenae is that of Aegeanized Achaeans, of an invading
race with northern affinities, who had adopted the arts of the Aegean
civilization, he considers that these finds prove that the earliest sauctuaries at
Olympia are prehistoric, and not post-mycenean. Apart from the question of
the Achaean origin of the pottery in question, this is very probable, but few-
will follow him in his revolutionary view that the 'Geometric' finds at
Olympia are pre- and not post-mycenean. When these views were first
formulated, after his earlier excavation at Olympia. they were vigorously
criticized by Furtwaenglcr. That some of his finds at Leukas. and
possibly at Olympia also, are altogether out of the Aegean context, and
are more related to the culture of central Europe, and oven that son]
them are earlier than Mycenean, is very probable, but that the 'Geometric

•' Marked I in PI. II of Schlieinami'a 'fir '■' Alh. Mitth. 1906, p. -

•296 R. M. DAWKINS

bronzes of Olympia are pre-mycenean is a view that it is impossible to
maintain. Similar bronzes have been found this year at the Spartau
sanctuary of Artemis Orthia, mixed with Geometric pottery, which immed-
iately underlies and gradually gives place to, not Mycenean, but Protocorinthian
and Corinthian pottery.

The excavation by which Dr. Doerpfeld claims to have found the site of
the Homeric Pylos, is near Zacharo, a little way north of Samikon. Three
Mycenean beehive-tombs have been discovered, and these, together with
the remains of a contemporary building, Dr. Doerpfeld, following the
'OfjLiipiKoorepoi of Strabo, 11 considers to mark the site of the Homeric Pylos.
The biggest of the tombs is thirteen yards (twelve metres) in diameter, with
a dromos about eight yards long cut in the rock. The crown of the vault of
the tholos, destroyed in antiquity, was perhaps as much as nearly forty feet
(twelve metres) high. After the burial the door of the tomb was closed with
a wall of small stones. In the floor of the tholos was a grave, covered with a
slab, but almost completely rifled. Better results however were obtained
from the tholos itself and the entrance, where many objects in amber, gold,
bronze, and ivory were found, besides bones and potsherds. The presence of
Roman graves above the debris shewed that the tomb was despoiled in
ancient times, and that these objects were only the remnants of its original
furniture. No pseudamphorae were found, and there were fragments of large
three-handled jars that recall the Cretan Palace Style. The period seems to
be that of the shaft-graves of Mycenae. It is of the greatest importance that,
whilst some of the bones shewed as usual no traces of fire, a number are
reported as being burnt.

On a hillock near by are walls, in which Doerpfeld recognizes an
Achaean building contemporary with the tomb. He gives it this name on the
strength of finding certain monochrome sherds like those of Leukas and
Olympia.

This practice of cremation points to the place being actually a very
early settlement of northerners, kinsmen to the later Achaeans of Homer, who
had assimilated the Aegean culture, but retained their own ancestral method
of disposing of the dead and partially their own style of pottery. The early
date however makes it clearly pre-homeric, a tomb possibly of the ancestors
of Nestor.

The excavations in Leukas, the Homeric Ithaka of Doerpfeld, have
been continued. In the summer of 1906 he dug in several places on the
island, notably the Nidri plain, where he locates the town of Odysseus, and
in a cave called Choirospilia. In both these places prehistoric sherds with
geometric patterns, either incised or painted, were found. This is again
his ' earliest Achaean ' pottery, found in all Achaean places, Leukas,
Olympia, and Pylos. He dissociates it altogether from the Mycenean, and
regards it us allied to the pottery of Halstatt and Villanova. It is difficult

1 'OuyipiKi&Ttpoi to?s tirtaiv b.Ko\ov6ovvT(% tovtov elval (f>u(TL tuv to? Nearopos TlvAov, uv ttji'
X*pav 5it£eimi' u 'AXtyetus. Strabo, 8. iii. 7.

ARCHAEOLOGY IN GREECE. 297

at present to know how much stress is to be laid on this 'Achaean pottery.
At Olympia he regards it as pre-niycenean, and makes it carry back with it
the geometric bronzes; at the Pylos settlement it is found with Mycenean
objects. He would perhaps regard it rather as national than as all of one
period. This is probably correct, only there seems no sound reason for
calling it Achaean.

To whatever criticism however Doerpfeld's views on Mycenean and
'geometric ' chronology may be open, he lias made it clear that one of the
most pressing problems of prehistoric Greek Archaeology is to clear up the
relations of these fabrics of north and west Greece, taking into account on
the one hand the neolithic pottery of Thessaly and the fabrics of central
Europe, and on the other determining their chronological position in the
sequence of the Aegean styles.

An important paper was read at the German Institute on the dis-
coveries which Dr. Prantl has made in working on the Parthenon sculptures.
By a careful study of the fragments he has made several new identifications.
For the east pediment the most important are the heel of Hephaistos, a
piece of the robe of Zeus, and two fragments of the wings of the Nike
In the west pediment he has identified the neck and part of the head of
Athena, a part of the body of Cecrops, and a great part of the figure of
Erichthonios.

Much progress has been made in excavating the great sites in Asia
Minor. The Ottoman Museum at Alabanda, the Austrians at Ephesus,
the Germans at Miletus and Pergamon, and the Danes in Rhodes are all
engaged on the work. Although the remains are generally of the Hellenistic
period, earlier and even archaic things are not lacking, and add greatly to
the interest of the results.

The chief work of the German excavation at Pergamon in the summer
of 190G was to clear the great gymnasium. This lies halfway up the hill,
and is on the highest of three terraces, on the lowest of which is the
gymnasium of the boys, and on the second that of the ephebes.

The results are extremely imposing. The gymnasium consists of a
large court, measuring 7S by 39 yards (70 by 35 metres). On the north
side the bases are all in situ, and it is proposed this year to replace as many
of the drums as possible. Along this side three fine halls have been cleared,
the westernmost of which had a Roman orchestra and auditorium built
over it.

Some houses in the town have been excavated, of which the 'Hoi
Attalos' contained interesting #aural paintings. It is intended ultimately
to clear all the lower town.

Not the least interesting part of the programme for the future is the
excavation of some tumuli, the largest of which. Jigma Tepe, is perhaps
undisturbed, and may contain Attalid tombs. A small tumulus that has
been dug contained a burial of the second century B.C. With the skeleton
were two swords and a gold crown formed of ivy leaves and decorated with
a figure e>f Nike.

298 R, M. DAWKINS

The latest results from the Austrian excavations at Ephesus in 1905 and
1906 are to appear in the next number of the Jahreshefte, of which Dr.
Heberdev has kindly let me see the manuscript, for the purposes of this
paper.

More pieces of slabs sculptured in low relief, and resembling a piece
previously discovered near the Library, have now been found. These slabs had
been moved from their original position to be re-used, and are fragmentary,
but it is clear that they once decorated an important building of the
Aurelian period. The nature and exact position of this monument remain
uncertain.

Between the Theatre and the Gate of Mithradates a large marble hall
in the Doric style with two naves has been excavated. The workmanship
points to the late Hellenistic period. The intercolumniations of the front
pillars were walled up later, possibly when an inscription running: Dianae
Ephesiae, Divo Clau[dio Imp. Neroni Caesari Augusto Germajnico, Agri[ppi]-
nae Aug[ustae, civitajti Ephesiorum, was cut on the architrave of the
southern wall. An earthquake in the reign of Tiberius may have necessitated
the restoration. The very late inscriptions on the walls prove that it
remained standing for many centuries. The longest of these is from the third
year of the Byzantine emperor Tiberius (a.d. 581). The bulk of it is in
Greek, but the last few lines, which give the date, are in Latin, cut by a
mason who evidently did not understand the language. A still later
inscription from the time of the emperor Heraclius and his son runs :
-f- 'HparcMijov) teal 'Hpa/c\i']ov twv OeocpvXdicTwv i)jjlo)v hecnroTwv, with the
addition in later lettering, /cat tcov irpacrlvaiv iroXka ra ery]-\-

South of this hall a circular building dating from the Greek period has
been found, and a late Roman rectangular hall, which may be the
avhenooptov, mentioned in an inscription published in the Jahreshefte vii.
Beiblatt, p. 52.

No work was done at the Church of the Virgin in 1900, but the as yet
unpublished results of 1905 are of some interest. 12 It occupied a space about
47<> by 105 feet, and consisted of three parts. In the west was a large rect-
angular court communicating by doors on the east with a church with narthex
and atrium. This had a nave and two narrow side aisles, and an apse at the
east end, and was roofed with a dome and barrel vaults. East of this again
was a second church with aisles, and to the south of this a baptistery.

Dr. Wiegand has kindly sent me the following notes on the work at
Miletus and Didyma in the season of 1906. At Miletus the Baths of the Empress
Faustina and the Lion Harbour have been "fully excavated, and between the
Delphinion and the Nymphaion a fine Hellenistic building with Propylon,
inner court, and halls has been found, that seems to have been the Prytaneion.
Belonging to a later date is an old Byzantine Basilica nearly 90 yards
(80 metres) long. It possesses an atrium, martyrium, and baptistery, and an
interesting circular plan and a mosaic floor. In the necropolis numerous

For an earlier report see Jahreshefte, viii, Beiblatt, p. 77.

ARCHAEOLOGY IN GREECE. 299

Hellenistic graves have been found. The work was much hindered by the
unusually heavy winter floods of the Maeander, and is to be continued this
autumn, with the special object of finding archaic remains.

Much has also been done to clear the temple of Apollo at Didyma, and
the re-erection of fallen blocks has greatly improved its appearance. The
systematic clearing of the Sacred Way has begun, and remains of several
archaic marble statues have been found. 13

The work of the Danish Archaeologists at Lindos in Rhodes conducted
by Dr. Kinch was finished in the summer of 1906. The main results have
been to clear the temple of Athena with its accompanying propylaea, portico,
and exedra. The temple, which replaced an earlier building, dates from the
fourth century. In this last campaign three important deposits were found
"ii the Acropolis of various statuettes, dating respectively from the seventh
or sixth, the end of the fifth, and the fourth century. A large number of
bases with artists' signatures were also found. There are as many as 114
examples and 04 different names. A fine relief of a ship cut in the ruck
was found near the entrance to the Acropolis, and in the neighbourhood
a rock-cut tomb with a two-storied facade. The upper of these is adorned
with four altars, and from its resemblance to a stage building Dr. Kinch
regards it of importance for the structure of the ancient theatre.

Lindos being now finished, Dr. Kinch is turning his attention to the
remains of a city that he has discovered at the south end of the island,
where almost all the visible remains, fragments of vases and statue!
belong to one and the same age, the epoch of the so-called 'Rhodian'
(Milesian) vases. This town is now to be excavated, and the early date gives
promise of extremely interesting results.

In a letter dated August of this year Dr. Kinch has very kindly
communicated the discovery of the necropolis of the town. The graves
he assigns to 800-600 B.C. No further details are yet to hand, but
good results may be confidently expected.

It remains for me to thank the archaeologists in charge of the various
excavations, who have so generously furnished the notes of their 1
results, from which this summary has been drawn up.

R. M. Dawkins.

J;! For notices of previous /.- •/'. Am. 1906.

MONEMVASIA.

Additional Notes.

Since the article on this subject (pp. 229 ff. above) was passed for
press I have obtained some additional information in Venice which it seems
desirable to put on record.

Of the two coats of arms illustrated on p. 240, that on the left belongs
to Sebastiano Renier. The other is of Antonio Garzoni, who was elected
podestd of Monemvasia in 1526, and again in 153.S, when he was the last
podestcl before the Turkish conquest. The identification of this latter coat I
owe to the assistance of Mr. H. F. Brown.

Hopf's statement, reproduced by me on p. 235, that Monemvasia was
Venetian in 1419, is, I find, not justified by the evidence. I have had some
difficulty in tracing the documents cited by Hopf, as the pagination of the
Vienna copy of the Midi used by him differs from that in the copy at Venice.
The three documents merely shew that Venetian wine-merchants were
engaged in the wine-trade at Monemvasia. They are as follows (I have
altered the Venetian dates to Modern Style) :

9 Jan. 1420.
Capta.

Attenta humili et devota supplicatione fidelium civiiuu nostrorum mercatorum

Monavaxie et Romanie et considerato quod mereantia huiusmodi vinorum hoc anno

parvum vol nichil valuit, ob quod ipsi mercatores multa et maxima damna sustinuerunt.

ol> quibus (sic) nullo modo possunt ad terniinum quatuor mensium sibi limitatum solvere

eorum datia prout nobis supplicaverunt ; Vadit pars quod ultra terniinum quatuor mensium

sibi concessum per terrain ad solvendum datia sua pro suis monavasiis et romahiis, conce-

datnr eisdem et prorogetur dictus terminus usque ad duos menses ultra predictos menses

quatuor sibi statuitos per terram ut supra dando plezariam ita bonam et sufficientem pro

ista prorogatione termini, qund commie nostrum sit securnm de datio suo, solvendo ad

termintiiu debitum.

De parte omnes.

(Archivio di Stato Venezia- Deliberazioni Senato Misti Reg. 53. c. 21.)

19 Feb. 1421.
< 'apta.

Quod audita devota supplicatione fidelium civium nostrorum mercatorum Romanie
it Monovasie Venetiis existentium, et intellectis damnis que receperunt iam annis tribus
de ipsis vinis et maxime hoc anno quia per piratas accepte sibi fuerunt plures vegetes
huiusmodi vim. rum, et considerato quod ilia que habent non possunt expedire, propter que

MONEMVASIA. 30 ]

damna nun possunt solvere sua datia ad terminum sibi limitatum per ordini El

audita superinde responsione offitialium nostrorum datii vini ex nunc captum =it quod
ultra dictum terminum sibi limitatum per ordines nostros elongetur terminus solvendi
dicta datia ipsorum vinorum usque duos alios menses.

De parte omnes.

De non 0.

Non sinceri 0.

(Archivio di Stato Venezia— Deliberazioni Senato Misti Reg. 53. c. 112.)

9 Feb. 1428.
In Consilio Rogatorum.
Capta.

Quod mercatoribus Monovaxie et Romanie, qui non potuerunt expedire vina sua
propter novitates presentes elongetur terminus solvendi datia sua per unum mensem ultra

terminum limitatum per ordines nostros.
De parte omnes alii.

De non 2.

Non sinceri 1.

(Archivio di Stato Venezia— Deliberazioni Senato Misti Reg. 56 carte. 761

Finally I note that the exact title of Mr. Papamichalopoulos' monograph
is: TloXiopKia teal aXcoatf ti)<; Movefi/3aaca<; vtto t&v 'EWijvoov tg3 1821.
laropiKt] irpwyfiareia virb Kcovar. N. Ha7ra/xL^a\o7rov\ov, Te\eio(f>oi,Tov

Ttjs VOflLKT)?. 'A@1)V7](Tl 1874.

W. Miller.

NOTICES OF BOOKS.

Essays and Addresses. By Sir Richaiid Jebb, O.M. Pp. vii+648. Cambridge:
University Press, 1907. 10s. 6d.

Our late President's occasional papers were invariably so carefully prepared that the
decision to publish a selection of them was amply justified. The present volume contains
the admirable study of Pindar and the monograph on Delos which appeared in the early
parts of this Journal. It also contains papers on ' The Genius of Sophocles,' 'The Age of
Pericles,' ' Ancient Organs of Public Opinion' (an address to Harvard students), ' Lucian,'
' The Speeches of Thucydides,' and ' Suidas on Sophocles and the Trilogy,' which deal
directly with Greek literature ; a review of Fronde's Caesar, the Rede Lecture on Erasmus,
the Romanes Lecture on Humanism in Education, and a Newnham lecture on Samuel
Johnson, which enter upon different fields of literary culture ; and five addresses dealing,
from various points of view, with the question of the position of University studies in
general, and classical studies in particular, with regard to modern life. The last of these
is the address to the British Association in South Africa on ' University Education and
National Life,' which was practically the author's last public utterance. The volume
as a whole is full of weighty substance, expressed with characteristic care and scholarly
moderation, and may be cordially recommended to the friends of classical studies.

Berliner Klassikertexte . Heft V., Griechische Dichterfragmente : Erste Halfte
epische und elegische Fragmente ; Zweite Halfte, lyrische und dramatische Frag-
mente. Bearbeitet von W. Schubart und U. von Wilamowitz-Moellexdorff.
Pp. viii -f- 136, with 2 facsimiles, and ii + 160 with 6 facsimiles. Berlin : Weidmannsche
Buchhandlung, 1007. 8 m. and 10 m.

The two parts which make up the fifth volume of the Berlin Kl issilcerte.de are a general
gathering of the minor verse fragments, on papyrus and vellum, in the Berlin Museum,
including several which have been previously published in periodicals. The first part
includes a list of Homeric papyri, a paraphrase of an Orphic poem on the rape of
Persephone, which is closely connected with the Homeric hj-mn to Demeter ; four frag-
ments of the Hesiodic Catalogue and a scrap of the Works and Days ; portions of Aratus
642-80?, 855-883, 922-934 ; Theocritus xi. 20-24, and xiv. 59-63, and a scrap of scholia ;
30 lines of Euphorion (the chief novelty of the part) ; 84 imperfect lines of a late
epic on the fortunes of the family and estate of Diomede during his absence before
Troy ; portions of 5 or 6 epigrams from a tiny roll, only two inches high ; two other
epigrams : Oppian, flalieut. v. 104-157 ; two fourth-century elegies on professors at
Berytus ; Nonnus, Dionys. xiv. 386-xvi. 30 (with lacunas); and three Byzantine
panegyrics, of the kind best known to us from Claudian. The second part is more
interesting, and contains, as its special novelty, considerable fragments of Corinna, in
strongly marked dialect ; 20 lines of one ode and 40 of another are in a good state
of preservation. In addition there are the Alcaeus and Sappho fragments which are
already known ; some well-written drinking songs, not much (if at all) later than 300 B.C. ;

NOTICES OF BOOKS 303

24 lines of the 'Axaia>i> criWoyot of Sophocles ; 52 lines of Euripides' Cretans, including
a vigorous rhetorical defence of herself by Pasiphae ; 35 choice lines from the Phaethon,
mutilated, but fortunately coinciding with the lines already preserved in the Codex
Claromontanus of the Pauline Epistles ; 50 lines of the Melanippi ; Hippol.VAZ 430, 193-
515, 816-624 ; Medea 513-560 ; Troades 876-879 ; Aristophanes, Ach. 598- 600, 631 633,
747-977 (with lacunas); Fro;/*, 234-262, 273-300, 404 410, 607-611; a few lettei
Birds, 819-829, 860-864 ; Clouds, 177-180, 207-20.), 233-235, 268-270, 936 972, and in
another MS. 955-988, 1007-1014: 30 imperfect lines from one play of the New Comedy
and 101 from another ; fragments of a Florilegium (including Eur. Hipp. 403-423) ; some
120 anapaestic monometers of the first century, of doubtful purport ; and a few un-
important scraps. The two parts contain a few interesting and important pieces (the
Sappho, Corinna, Cretans, and Euphorion), and the publication of the rest at leasl
consciences of the curators of the Berlin Museum.

Etude sur Didymos, d'apres un papyrus de Berlin. By P. Foucart. Paris
(extrait des Memoires de l'Academie des Inscriptions et Belles-Lettres, tome xxxviii,
1907). Pp. 193. 7 f. 50 c.

This is an important study of the commentary of Didymus on the Philippics, originally
published in the Berliner Klassikertexte, vol. i. The conclusions of Foucart with regard to
the character of Didymus and his compilations deserve to be studied in connexion with
those of Diels ; and the second part, which contains separate studies of all the copious
quotations from other authors (especially Philochorus), supplies most usefully a deficiency
in the original publication.

Das Buch bei den Griechen und Romern. By W. S hubart. Pp. 159; U
illustrations in text. Berlin: Handbiicher der kgl. Museen, 1907. 3 m.

A very useful summary of present knowledge on the subject of Buchwesen in antiquity,
written in a clear and popular style, without foot-notes or references. The information is
very complete so far as it goes, and the author (who is curator of the Greek papyri in the
Berlin Museum) has made good use of his opportunities of bein^ acquainted with the
results of modern discoveries. The four chapters deal with ' Writing Materials,' ' th<_-
Roll,' 'the Codex,' and 'Copying and the Book-trade.'

Greek Papyri in the British Museum : Catalogue, with texts, vol. III. Edited by
F. G. Kenvon and H. I. Bell. Pp. lxxiv + 388. £2 10s. Atlas of facsimiles to the

above (100 plates). London, 1907. £3 3s.

The third volume of the British Museum Catalogue of Papyri i.- on the same lines as its
predecessois. It contains a numerical catalogue of S40 papyri, and texts (with briel
introductions and notes) of 248 (all non-literary documents). These are divided into
chronological groups as Ptolemaic, Roman, Early Byzantine, and Late Byzantine, with
subdivisions in each group, except the first, according to subject. The texts are "i the
usual kind. They include some well-preserved Ptolemaic contract.- from Pathyris, some
long land-registers of the first century, the account- of the waterworks-commissioners ot
some town (probably Arsinoe or Hermopolis), a brief narrative of a voyage up the Nile,
and a diploma of membership in an athletic club, granted to a boxer at the great games at
Maples in a.d. 192. The indices are on the usual full scale. The atlas of facsimiles
provides a series of 100 plates, most of them precisely dated, and ranging from L9S B
711 A.D.

304 NOTICES OF BOOKS

The Tebtunis Papyri. Part II. Edited by B. P. Grenfell and A. S. Hunt, with
the assistance of E. J. Goodspeed. (University of California publications, Graeco-
Roman archaeology, vol. II.) Pp. xv + 485, with a map and 2 plates. London and
Xew York, 1907. £2 5s. net.

The second volume of the Tebtunis papyri (discovered and edited by Messrs. Grenfell and
Hunt on behalf of the University of California) has appeared, through the munificence of
Mrs. Hearst, without being delayed by the disaster which befell the University in the
recent earthquake. Its interest is mainly for the regular student of papyri. It contains
only four literary texts (apart from small fragments). Two of these are from Homer (II. ii
and xi), and one from Demosthenes (De Fals. Leg.) ; the fourth, and most interesting, is
a portion of the lost Greek original of Dictys Cretensis. This, being in a hand of the early
third century, proves that the work must have been composed not later than the second
century, and possibly earlier. The documentary texts are of the usual miscellaneous kind,
excellently edited, and provided with full indices. There is a long appendix on the
topography of the Arsino'ite nome, with a map.

Die Griechische Skulptur. Von R. Kekule' von Stradonitz. [Handbiicher der
koniglichen Museen zu Berlin.] Pp. iv + 383; 155 illustrations. Berlin: Reimer,
1906.

This work is a manual of Greek sculpture, described in historical sequence, and is
primarily intended for the use of visitors to the Berlin collections of original sculptures
and of casts. For the purposes of the book, Greek sculpture begins with the primitive
examples from the Athenian Acropolis, and closes with examples of Roman imperial
architecture, and sarcophagi. In accordance with the scheme and object of the writer, the
chief stress is laid on those sections which can be illustrated by original works at Berlin,
and the literary side of the history of sculpture is only introduced as far as is necessary.
The work could, however, be used to a considerable extent, as a companion in any gallery
of casts.

A Handbook of Greek Sculpture. By Ernest Arthur Gardner. [Handbooks of
Archaeology and Antiquities.] Pp. xxxiv + 591. London (Macmillan) and New
York (The Macmillan Co.), 1907. 10s.

In this reissue of the revised edition of Prof. Gardner's handbook (see J.H.S. xxvi. p. 183)
it is satisfactory to be able to say that the illustrations, the condition of which previously
left something to be desired, have been thoroughly overhauled, and are now worthy of the
text. An outline reproduction of Prof. Furtwangler's reconstruction of the Aegina
pediments serves as frontispiece.

Roman Sculpture from Augustus to Constantine. By Mrs. Arthur Strong,
LL.D. Pp. xx + 408. 130 plates. London : Duckworth and Co., 1907. 10s. 6d.

Mrs. Strong has made a courageous attempt to deal in a <mall compass with the great mass
of Roman sculpture from Augustus to Constantine, and has performed a service to
archaeologists and particularly to student- and teachers of Roman history and civilisation.
Her book contains by far the most complete enumeration of Roman monuments in the
English tongue, and a great number of plates, on the whole admirably reproduced. In
addition, Mrs. Strongs wide acquaintance with continental archaeological literature
provides in the notes references to the most important treatises dealing with the particular

NOTICES OF BOOKS

monuments. Of these the reliefs are in many ways the most important remains, and at
the same time the least accessible, Mrs. Strong's careful description and full illustration
of the three great historical documents in relief, the Ara Pacis Augusti, the Col imn of
Trajan, and the Column of Marcus Aurelius, alone make her work indispensable to
English students of Roman history^ For the treatment of all three she mainly reli<
Petersen^ careful studies, and the description of the Ara Pacis is as complete a, may he,
until the innumerable fragments in the Museo delle Terme are pieced together and
described. The Column of Trajan is illustrated by twelve plates and a running com-
mentary on the whole, the Column of Marcus Aurelius by seven plates. In chronological
relation to these, Mrs. Strong reproduces and discusses the remaining important reliefs, the
panels of the arch of Titus, the arch of Trajan at Benevento, the Hadrianic and Aurelian
reliefs in the Palazzo dei Conservatory and the whole complex of heterogeneous panels
which decorate the arch erected by a grateful Rome to Constantine. Mrs. Strong has
selected wisely both from the vast mass of monuments at her disposal and from the no
less vast abundance of ideas thrown out by such writers as Wickhoff and Riegl. The
reliefs of the arch of Titus, on which as a basis Wickhoff founded his theory of the
independence and pre-eminence of Roman Art, receive, on the technical side, at least a
due appreciation ; and Riegl's bold advocacy of neglected virtues in the 4th century
friezes on the Constantinian arch is briefly expounded. The last chapter, on Roman
Portraiture from Augustus to Constantine, is the least satisfactory. The illustrations,
which include three plates of Imperial coins, are good and reproduce less known and finer
busts. The text, however, attempts to deal with the Imperial busts in general, a task too
great for the space allowed.

Nouvelles Archives des missions scientifiques et litteraires. Tome xiii.
fasc. 4. Une Forteresse Iberique a Osuna (Fondles de 1903. Par MM. Arthur
Ex.. el et Pierue Paris. Pp. 131, with 40 Plates. Paris: Imprimerie Nationale,
1906.

This well-illustrated number of the Nouvelles Archives contains the results of the
excavation of the older portion of the site of the antique Urso, the modern Osuna in
Southern Spain. The existence of Roman remains there has long been known, but the
stimulus to the French archaeologists was the hope of the discovery of the enigmatical
pre-Roman Iberian work. A portion of a hastily built fortress, which can be dat
numerous sling glands inscribed with CN. MAG. IMP., was dismembered and contained
sculptured fragments of earlier date. These fragments and the rinds generally are carefully
tabulated and reproduced. The work presents that known admixture of archaic f
and incompetent execution which inevitably suggests the local imitation of imported
models and the character of the forms permits a comparison with prototypes of the
various dates and sources, early Greek and Phoenician, even Chaldaean and Mycenaean.
Two tombs found on the site but earlier than the fortress contained remains which by
comparison with articles found in Punic tombs at Carthage are inferred to 1 nic

L'art Greco-Bouddhique du Gandhara. By A. Fou< mi;. Vol i. Pp. xii-f-

300 illustrations, 1 plate, and a map of the Gandhara district. Paris: Imprimerie
Nationale, 1905.

This admirable work, when completed, will constitute much the most exhaustive account
yet undertaken of the Buddhist art of the N.W. provinces of India, an art bo interesting
from the extraordinary flexibility shewn by the Grae< (-Roman mould in adapting itself to
absolutely alien modes of thought. This volume deals with the buildings and bas-reliefs,
so far as the latter are accessible ; vol. ii is to contain the mably including

U.S. — VOL. XXVII. X

306 NOTICES OF BOOKS

Dr. Stein's new material for their study. The author takes the traditional life of Buddha.
step by step, tracing each act or scene in the monuments,— a great advance on anything yet
done ; while the sections in which he disentangles the classical and native elements in the
art of Gandhara, and analyses the two waves of western influence that, at separate times,
reached India, shew the skill and sobriety of a master. That old problem, the Mathura
sculptures, troubles no longer : these now fall into their place as imitations of Gandhara
art, as indeed Mr. Vincent Smith has also recently seen. The question of the derivation of
the classical influences in the Gandhara school is, however, so lightly treated as to suggest
a fuller discussion to come, unless M. Foucher has nothing to add to the brilliant essay in
which be put forward his theory of ' wandering artists ' thirteen years ago. As to date, he
follows Senart in placing the inferior limit for the main outburst of this art not later than
the middle of the second century a.d., on the strength of the Amravall inscriptions. But
the superior limit fluctuates with Kanishka. He appears to favour the end of the first
century a.d., and two or three years ago it seemed tolerably safe to connect the beginnings
of the Gandhara school with the great peace of Hadrian and the Antonines ; but now,
should Dr. Fleet be right in referring Kanishka's date to the Vikrama era (58 B.C.), one
may have again to reckon with the possibility of a much earlier commencement. In any
case, there is" a remarkable resemblance between these long acts of the Buddha, unwinding
themselves round the base of some stupa, and Trajan's column. If one grumble at this
splendid book be permissible, it is over its title. M. Foucher could, if he would, have
killed off the term ' Graeco-Buddhist,' with its misleading associations. Now it is too late.

Ancient Khotan. By M. Aurel Stein. Vol. i. Pp. xxiv + 621, and 72 illus-
trations. Vol. ii. 119 plates, and a map of Khotan. Oxford : Clarendon Press,
1907.

These magnificent volumes contain the detailed report of the archaeological explorations in
Chinese Turkestan carried out by Dr. Stein in 1900-1 on behalf of the Indian Government,
together with appendices by leading experts in various branches of Oriental research.
The main outline of Dr: Stein's discoveries at the desert sites excavated by him is already
familiar from his Preliminary Report (1901) and his personal narrative, 'Sand-buried
ruins of Khotan' (1903); it is now possible to disengage the results of interest to the
classical student. The outstanding fact is the influence, in the early centuries a.d., of
India, and more especially of the ' Graeco-Buddhist ' art of Gandhara, on the art of
Khotan, through which channel smne classical forms, e.g. the arrangement of the drapery on
the standing Buddha- figure, reached Japan ; but Khotan throws no light on the origin
of the Gandhara school itself. The celebrated classical seals on wooden KharosthT
documents of the third century a.d., and the classical intaglios from Moji, are now
definitely called Rom m work of the third or possibly of the second century a.d. A restored
enlargement by Mr. F. H. Andrews of one of the figures of Athene Promaehos, with aegis
and thunderbolt, adorns the title-page ; but perhaps even more striking than the juxta-
position of Roman and Chinese seals on records from the same office is the resemblance of
the Nagini of shrine D. ii at Dandan-Uiliq to the Venus de' Medici. Doubtless the seals
came in by the great trade route to China, described by Marinus, as to which Dr. Stein
now abandons his former adherence to the identification of Ptolemy's Xldtvos nvpyos with
Tashkurgan. Iranian artistic influences, as well as speech, are seen extending to Khotan ;
of Greek influence proper, unless filtered through India, no trace appears. Whether any
or what elements at Khotan came from Bactria we cannot say, as we do not know what
Graeco-Bactrian civilisation was like. If only Dr. Stein could excavate Balkh '.

NOTICES OF BOOKS 307

Quelques Reformes de Solon. Essai de critique historique. Par Chablbs Gilliark.
Lausanne : G. Bridel & Cie, 1907. Pp. 324.

M. Gilliard deals principally with the economical and financial reforms. In the extant
fragments of Solon's own poems this part of his work is very prominent, and tin-.
fragments are the most reliable evidence lor the life and work of their author : the ancients
did not understand their middle ages as we do ours. A considerable part of the book is
then devoted to an account of the economic troubles in pre-Solonian Athens, with
chapters on such subjects as the Rich, the Poor (ne'Kdrai, $?)T fS , Ur^^pni, etc), Debt and
Mortgage (with a discussion of the Solonian Spot). Solon's reforms are dealt with m
such main headings as the Law against seizing the Debtor's Person, The Right of Bequest,
the Seisachtheia, the Solonian Classes, Solon's Monetary Reforms. One chapter before the
end discusses Solon's political reforms. The book contains a complete collection of
Solon's poems, a bibliography, and an excellent table of contents.

Thucydides Mythistoricus. By F. M. Cornford. Pp. xvi+252. London :
Arnold, 1907. 10s. 6f/.

Mr. Cornford's book is divided into two parts. The first ('Thucydides Historicus') deals
with what he regards as the inadequate explanation given by the historian of the cause of
the Peloponnesian War. It was, he holds, really brought on by the commercial party who
dictated the policy of Pericles. The Megarian decrees were the first step in this policy,
and were part and parcel of the design to secure Allien ian trade with the West at the
expense of Corinth. All this Thucydides more or less ignores, because he has no true
conception of social and economic conditions, in which we better instructed moderns find
the explanation of historical events. The second part ('Thucydides Mythicus') attempts
to shew how the historian's mind was — as was inevitable in his age — dominated by
conceptions such as Tyche, Elpis, Eros, etc., which are for him not mere abstraction--,
names given by men to the Tarious forms in which natural l«w expresses itself, hut
living agencies. This attitude of mind leads the historian to arrange and mould the
materials of history much as a dramatic poet of the time would deal with them. The
mythical atmosphere causes the actual facts to be unconsciously omitted or from the
modern and moro privileged point of view, distorted. Mr Cornford has written a most
brilliant essay, but cannot be said to have penetrated below the surface of his subject
His acquaintance with previous writers thereon, or with the archaeological questions that
are occasionally involved, leaves a good deal to be desired. But that is a minor objection.
He proceeds on the somewhat naive assumption that wars are sufficiently explained by
economic motives. Whereas no questions of trade will bring two stales into conflict unless
the psychological conditions are favourable ; and these condiiionsare adequate y recognised
by Thucydides. It is also difficult to believe that Thucydides was so dense, or mi little in
the confidence of the politicians of his time, as .Mr. Cornford's theory presupposes. As
regards the method of the book, Mr. Cornford forgets that, once it is assumed that the real
meaning of an author's words is something very different from what he thinks he is saving, the
critic's interpretation becomes purely subjective ; it becomes, too, interesting not becai
the light which it seems to throw on his Bubject, but because of the attitude of mind
which it reveals in the critic himself. From the purely literary point of view it may be
justifiable to make an unoffending historian the object of a psychological jeu d? esprit, and
to treat his perfectly reasonable versions of almost contemporary events — such a^ the
story of Harmodius and Aristogeiton — as episodes in a work of die imagination. But for
the sake of the advance of historical criticism, the habit is to be deprecated. Xaptom pei
8!) raiira, "KUiv be 8(ivoi>.

X _'

303 NOTICES OF BOOKS

The Second Athenian Confederacy. By F. H. Marshall [Cambridge Historical
Essays, No. XIII]. Pp. x + 136. Cambridge University Press, 1905.

We have delayed too long to notice Mr. Marshall's little book, which, unlike most Prize
Essays, is a solid and permanent contribution to the study ol its subject. So much has
been discovered in the form of inscriptions since the publication of Busolt's monograph
more than thirty years ago that, pending the completion of his history, one is grateful for
a careful and lucid summary of the present state of our knowledge of this difficult
transitional period. For the histories of Holm and Beloch are either too sketchy or too
tendenzius to be thoroughly satisfactory. Mr. Marshall's first chapter gives a brief
description of the events leading up to the foundation of the confederacy, of which the
most important was the formation of an anti-Spartan league in 394-3 B.C. It is well known
that the evidence for this league is chiefly numismatic ; but a coin of Byzantium, which was
unpublished when Mr. Marshall was writing, has introduced an element of difficulty into
the chronology. This coin, which is uniform with those of the other members of the league,
shews that Byzantium belonged to it ; but whereas Ephesus, Samos, and Cnidus joined
Sparta and, in Mr. Marshall's opinion, broke up the league in 391, it is improbable that
Byzantium can have joined it before the expedition of Thrasybulus in 389. Chapter II deals
with the general principles of the Confederacy, the remainder of the book describes its
history, which is one of decline from the very beginning. The economic historian would
attribute this melancholy fact to the imperfect financial organisation of the league ; but
that was only a symptom of the real causes. These lay first in the inability of Athens to
regard herself as merely the chief member of the league, and not its mistress ; and second
in the rise, in Macedonia, of a power which was hampered neither by any scruples as to
the subjection of other powers, nor by lack of men or money to carry its policy of con-
quest into effect. Mr. Marshall rightly regards the Confederacy as the forerunner of true
federation in Greece, but it is equally significant as the last futile appearance of the city-
state as an imperial power.

Sources for Greek History between the Persian and Peloponnesian
Wars. Collected and arranged by G. F. Hill. Second Issue. Pp. xii + 439.
Oxford : Clarendon Press, 1907. lO.s. Qd. net.

In this re-issue an Appendix of seventeen pages of newly acquired material and corrigenda
has been added and the list of aichons revised. Otherwise the book is the same as it was
when first published in 1897.

The Greatness and Decline of Rome. Volume I. The Empire Builders : Volume
II. Julius Caesar. By Guglielmo Ferrero ; translated by Alfred E. Zimmern.
Pp. vi + 328, vi + 389. London : Heinemann, 1907. 17s. net,

These two volumes contain a history of the age of Caesar from the death of Sulla to the
Ides of March, with five introductory chapters giving a broad outline of the four preceding
centuries. Though complete in themselves, they do not conclude Professor Ferrero's task.
His intention is to continue the narrative, in succeeding volumes, down to the break-up of
the Empire, and references in the work now published suggest the early appearance of a
further volume.

The detailed narrative traverses a much studied generation, presenting not a few of the
familiar questions in an unfamiliar light. The characters are drawn with great skill and
insight. Tin.- political adventurer Vatinius, Crassus the aspiring financier, Clodius the tur-
bulent demagogue, and the varied crowd of politicians and generals that pass across the stage,
all are portrayed witli life-like vigour : in particular the picture of the old campaigning
aristocrat Lucullus is intensely vivid. In estimating the political aims and powers both of
Pompeins and of Cicero, the author has pursued the via media; Cicero especially is
given his due, while the author's evident divergence from Mommsen has not led him into

NOTICES OF BOOKS 309

undue eulogy. But of necessity the character which dominates the work is that of Julius
Caesar, and here the devout follower of Mommsen will be disappointed or disillusioned.
( laesar'a early career, we tind, was largely determined by the pressure of his debts. The
Helvetian campaign was a cardinal mistake, barely redeemed by the expedition against
Ariovistus. A brilliant general and a brilliant opportunist, Caesar appears in the end not
as a state-man but as an arch-destroyer unable to restore the fabric which he had helped to
shatter. Though this view of the true Julius Caesar is developed and elaborated with
great ingenuity, it would be difficult to accept it without reserve.

While the book contains many fine pen-portraits, it is less a history of men than of
movements. It is above all noteworthy for its correlation of social with political history.
The condition of the capital, of Italy, and of the Empire, finds its proper place as a key
to the shifting and almost bewildering schemes of parties ; and the prevalence of the slave
traffic, the temporary influx of wealth, and the long-continuing question of debt are rightly
emphasised. The treatment of the social questions never fails to be interesting, for the
author writes in an exhilarating style which hurries the reader along. But, partly no
doubt on account of the difficulty and obscurity of the subject, it appears sometimes to be
repetitive and circumscribed. In the history of sociabmovements, the notes here and there
confess to conjectures in the text, and in other cases the reader looks in vain for any sub-
stantial authority cited in support of the author's views. Professor Ferrero denies to his
protagonists something of the sagacity and power in moulding events which other historians
concede to them, and. lays greater stress on the broad unconscious movements of the times.
Probably in doing so he attains a truer perspective, but he essays a more difficult task.
The result is a work which is always stimulating but not always convincing. Still, while
suggesting that the final word on the age of Caesar remains unspoken, the work undoubtedly
constitutes a valuable addition to existing knowledge of the period.

As is natural in so long a book, points of detail suggest themselves for criticism. For
instance, the dating of the Lex Aebutia circa 100 b. c. will hardly find general acceptance.

Mr. Zimmern has carried out the translation with singular felicity.

Studies in Roman History. By E. <i. Hardy, M. A., D. Litt. Pp. viii + 349.

London : Swan Sonnerischein and Co., 190<>. 6s.
In this book Mr. Hardy republishes his ten Essays on ' Christianity and the Roman Govern-
ment,' and adds six miscellaneous studies touching on aspects of imperial history. The essaj -
on Christianity give a judicious and compendious account of the attitude of the Government
to the new religion from its first appearance in the Empire to the time of the Antonines.
A subject often viewed in the spirit of partisanship or prejudice is here treated in the sober
light of scholarship, and Mr. Hardy's well-reasoned conclusions leave an impression of the
remarkable forbearance of the Government towards a movement which it did not, and
could not, understand,— a forbearance which contrasts strangely with the history of similar
relations in other epochs.

One of the further studies which now first appear in the book deals with the little-
known provincial assemblies, of which the primary object was the regulation of the worship
of 'Rome and the reigning Emperor,' and which kept up in the provincial populations the
sense of their connexion with, and dependence upon, the imperial city. Mr. Hardy s study
is of exceptional value for the side lights which it throws on the dark questions of provin-
cial organisation.

Other studies deal with the constitution of the army in the time of Augustus, the
movements of the legions in the first two centuries, a Bodleian MS. of Pliny, and other
detached questions. They are very technical and their interest very specialised : while
they add something of value to tin mass of research, they rather spoil the uniformity of .Mr.
Hardy's book. The first ten studies form a consecutive narrative which will be generally
read with interest and profit ; the remaining essays form a miscellany in which only the
expert will find much interest.

310 NOTICES OF BOOKS

The Stoic Creed. By William L. Davidson. Pp. vi -f 274. Edinburgh : T. and T
Clark, 1907. 4s. (id.

The public whom Professor Davidson has in view is not any limited circle of specialists
but the larger world of general culture which may be concerned to know the broad features
of Stoicism and the bearings of Greek philosophy on modern life. If therefore one must
pronounce that to the serious student of philosophy or classical antiquity the book is not of
much use, this is not to deny that it will have its use in the sphere for which it was
intended. It is perhaps to look at it from a somewhat narrow point of view, although that
may be appropriate to a journal devoted to Hellenic studies. It would indeed be unreason-
able to complain of its not being what it does not profess to be — a monograph intended to
shed new light upon the darker parts of its subject, or a handbook which might guide
beginners over the field. Since it is neither, there was no obligation to indicate the special
literature on Stoicism. But it is not that important works which have appeared in the
last twenty years are not mentioned (by Bonhoffer, Schmekel, Dyroff, Aal) : the suspicion
is forced upon one that Professor Davidson has not himself taken the trouble to become
acquainted with them. His Zeller, of course, he knows, and he mentions Stein's ' Psycho-
logie' (Vol. II, 1888) — the most unsafe guide, by the way, to whom any one taking up
Stoicism could be referred. The careful and elaborate study of the Stoical philosophy,
contained in Bonhoffer's two volumes on Epictetus (1890, 1894), put many points in a
wholly new light. Professor Davidson repeats old suppositions as if they had never been
questioned — the view for instance that all knowledge, according to Stoicism, had a sensuous
origin. The idea again that the goodness of the Deity was not recognized by the Older
Stoics (pp. 60, 90) is certainly wrong (cf. Chrysippus ap. Plut. de Stoic, rep. 105 le). The
account of KadfJKov, with the statement 'that it is applicable only to things indifferent'
(p. 154) is very unsatisfactory. Professor Davidson's terminology is sometimes curious.
' Affects,' for instance (p. 49), as a translation of nddr], though familiar in German, is new to
me in English, and does not seem to correspond with the technical meaning of the term
given in Baldwin's 'Dictionary of Philosophy.' 'Artificial fire' (p. 88) will hardly do as a
translation for nvp rex^^ov ; can Professor Davidson have been thinking oifeux d 1 artifice?
1 Homocentric' (p. 183) is not the same as 'anthropocentric' : it does no

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "The Journal of Hellenic studies"

See other formats